an improved mouldless manufacturing method for …€¦ · value analysis (pva). subsequently, the...
TRANSCRIPT
![Page 1: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/1.jpg)
AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR FOAM-CORE COMPOSITE SANDWICH STRUCTURES
By
Mario Mahendran
A thesis submitted to The Faculty of Graduate Studies and Research
in partial fulfillment of the degree requirements of Master of Applied Science
Ottawa-Carleton Institute for Mechanical and Aerospace Engineering
Department of Mechanical and Aerospace Engineering Carleton University
Ottawa, Ontario, Canada December 2010
Copyright © 2010 -M. Mahendran
![Page 2: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/2.jpg)
1*1 Library and Archives Canada
Published Heritage Branch
395 Wellington Street OttawaONK1A0N4 Canada
Bibliotheque et Archives Canada
Direction du Patrimoine de I'edition
395, rue Wellington OttawaONK1A0N4 Canada
Your file Votre r$f6rence ISBN: 978-0-494-79537-8 Our file Notre reference ISBN: 978-0-494-79537-8
NOTICE:
The author has granted a nonexclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or noncommercial purposes, in microform, paper, electronic and/or any other formats.
AVIS:
L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'lnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats.
The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.
L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation.
In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis.
Conformement a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these.
While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis.
Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant.
14-1
Canada
![Page 3: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/3.jpg)
ABSTRACT
An improved mouldless Liquid Composite Moulding (LCM) manufacturing method for foam-
core sandwich composite components was developed and utilized to manufacture a full scale
fuselage for the GeoSurv II Unmanned Aircraft System (UAS). Implemented as part of the low-
cost composite airframe research at Carleton University, this work intends to improve upon the
previously implemented mouldless Vacuum Assisted Resin Transfer Moulding (VARTM), using
effective design and manufacturing methods. Such methods will directly benefit small
aerospace companies, especially those dealing with general aviation aircraft and UASs.
In this work, numerous state-of-the-art LCM processes were reviewed. These identified Closed
Cavity Bag Moulding (CCBM) as a potential alternative to VARTM for mouldless manufacturing.
A series of experiments was carried out on CCBM process to evaluate its feasibility, including
investigations of various bag sealing and resin distribution strategies, followed by a Process
Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved
manufacturability. The new design was optimized using finite element analysis in Abaqus.
Finally, a full scale GeoSurv II fuselage was manufactured to demonstrate the viability of the
developed process. Results showed that this process is a viable option for manufacturing
complex foam-core composite components in small quantities. Improved part quality,
tolerances and weight reduction of 36% were achieved using the optimized design and
manufacturing method.
ABSTRACT ii
![Page 4: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/4.jpg)
DEDICATION
Dedicated to my Mummy, Mary Rosaliya Mahendran and Dada, Kanesapillai
Julius Mahendran for their unconditional love and support.
DEDICATION i i i
![Page 5: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/5.jpg)
ACKNOWLEDGMENTS
First of all, I would like to thank my supervisors, Prof. Paul V. Straznicky and Prof. Jeremy Laliberte for giving me this research opportunity and offering continuous guidance. As much as I appreciate their expertise, I must thank them for their exceptional patience throughout the project.
I would like to acknowledge Sander Geophysics Ltd. (SGL), NSERC Collaborative Research and Development (CRD) Grant and NRC-IAR for providing the financial resources for this research.
I also extend my appreciation to research officers, Dr. Chun Li, Drazen Djokic and Dr.Peter Krimbalis at the National Research Council-Institute for Aerospace Research (NRC-IAR) for providing consultation and support throughout my Master's program. I must also thank Tom Kay at NRC-IAR for his support in cutting the fibreglass rods.
My sincere gratitude goes to the technical staff of the Mechanical and Aerospace Engineering Department at Carleton University: to Alex Proctor for his support with machining of the foam cores, to Kevin Sangster for his guidance in the machine shop and to Steve Truttmann for his useful insights and support in the structures lab.
I also thank Aaron Miller (Composites Canada), John Biron (DIAB Inc.), Larry Audette (Prairie Technology Group Inc.) and Dave Kindt (Kindt-Collins Company LLC) for their interesting technical insights and support to this project. My appreciation also goes to Michel Reid, at Progress Plastics and Compounds Inc. for providing me with the opportunity to participate in the CCBM demonstration.
I am truly thankful to Frank Cappelli, Eileen Ruddek and Russ Elkin at 3A Composites/Baltek Inc., for their kind support and material donation.
Some of the experimental work in this thesis was completed with the assistance of several undergraduate and graduate students of the GeoSurv II UAS project. I am particularly thankful to Quinn Murphy (undergraduate summer student May-Aug. 2009) for his assistance with mechanical testing and documentation, Alexandre Adcock and Mauricio Buschinelli (GeoSurv II UAS project-2009/2010) for their support with the fuselage redesign work, Alan Lares (continuing graduate student) for his assistance with mechanical testing and fuselage manufacturing, Shashank Pant (continuing graduate student), who was always present when an extra hand was needed, Jeff Teutsch and Jerry Mac Pherson (GeoSurv II UAS project-2010/2011) for their assistance with characterizing the dimensions and quality of the fuselage. I am also thankful to Laurent Loisy, an exchange student from France, who provided valuable help with the final fuselage manufacturing.
My heartfelt gratitude goes to my parents and all members of my family for their love, care and support. Success would have been far from reach, without your role.
Special mention needs to be made for the wonderful human beings, who, through invaluable friendship, kept nourishing my life, both inside and outside of Carleton. Specifically:
• Jeeva, for all the wonderful moments we encountered at Carleton. You truly are a pillar of success in my academic career. Thanks for being around whenever needed, pointing me towards the right direction.
• Shashank and Fady, for the memorable coffee times and wing nights, where many exciting 'principles of life' were exchanged. My only regret is not getting to know you guys earlier in my career.
• Masih & Majed for being present 24/7 in the lab, exchanging ideas & sharing stories of life. • Dave, Henry and all other fellow graduate students in 2350 & Grad Lab • Mech & Aero Crew and others who stopped by the lab every now and then. • Sharmi & Co. for all the Quality Time. I won't thank you guys, but here, I'll say this one time 'you are my best gift'. • Not to be forgotten, I thank Nirth and Van for accompanying me in some labour intensive manufacturing tasks,
without a shadow of hesitation.
Last but not least, big thanks to everyone who periodically questioned the status of my thesis. Bluffing those questions was a definite inspiration, when intruded by other intellectual aspects of life.
ACKNOWLEDGEMENTS IV
![Page 6: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/6.jpg)
CONTENTS
ABSTRACT ii
DEDICATION iii
ACKNOWLEDGMENTS iv
CONTENTS v
LIST OF TABLES ix
LIST OF FIGURES x
CHAPTER 1. INTRODUCTION 1
1.1. Background 1
1.2. Overview of the Low-Cost Composite Airframe Research 2
1.3. Overview of the GeoSurv II UAS 4
1.4. Thesis Objectives and Organization 6
1.5. Contributions 8
CHAPTER2. OVERVIEW OF THE GEOSURV II FUSELAGE 10
2.1. Design of the GeoSurv II fuselage 10
2.2. Manufacturing of the Fuselage Prototype 12
2.3. Post Process Assessment of the GeoSurv II Fuselage 13
2.4. Design and Manufacturing Objectives for the New Fuselage 15
CHAPTER 3. REVIEW OF LCM PROCESSES 17
3.1. LCM Processes 17
3.1.1. Resin Transfer Moulding (RTM) 20
3.1.2. Structural Reaction Injection Moulding (SRIM) 21
3.1.3. Vacuum Assisted Resin Transfer Moulding (VARTM) 22
3.1.4. Seemann Composite Resin Infusion Moulding Process (SCRIMP™) 26
3.1.5. Vacuum Assisted Process (VAP®) 26
3.1.6. Fast Remotely Actuated Channelling (FASTRAC) 27
3.1.7. Controlled Atmospheric Pressure Resin Infusion (CAPRI) 28
3.1.8. Double Bag VARTM 29
3.1.9. Advanced VARTM (A-VARTM) 30
3.1.10. Single Line Injection (SLI) 31
3.1.11. High Performance VARTM (Hyper-VARTM™/ Hyper-RTM™) 32
3.1.12. Vibration Assisted Liquid Composite Moulding 33
CONTENTS v
![Page 7: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/7.jpg)
3.1.13. Closed Cavity Bag Moulding (CCBM) 33
3.1.14. Co-Injection Resin Transfer Moulding (CIRTM) 34
3.1.15. Euro Composites® Honeycomb Liquid Moulding (EC-HLM) 35
3.1.16. VacFlo® Process 37
3.1.17. Light RTM 37
3.1.18. Resin Infusion between Double Flexible Tooling (RIDFT) 38
3.1.19. Flexible Injection 40
3.1.20. Resin Film Infusion (RFI) 41
3.1.21. Semi-Preg Infusion 42
3.2. Suitability LCM Processes for Mouldless Manufacturing 43
CHAPTER 4. CCBM PROCESS DEVELOPMENT 46
4.1. Mouldless CCBM manufacturing considerations 46
4.1.1. Sealing Mechanisms for Mouldless CCBM 46
4.1.2. Resin Inlets and Outlets 47
4.1.3. Resin Distribution in CCBM 48
4.2. CCBM Manufacturing Trials 48
4.2.1. CCBM Manufacturing Trial # 1 49
4.2.2. CCBM Manufacturing Trial #2 51
4.2.3. CCBM Manufacturing Trial # 3 52
4.3. Value Analysis of the CCBM process 53
4.3.1. Introduction to Value Analysis 54
4.3.2. CCBM Process Value Analysis 55
4.3.3. PVA Results 56
4.3.4. PVA Conclusions 57
CHAPTER 5. FUSELAGE MATERIAL SELECTION 58
5.1. Sandwich Theory 58
5.2. Core Material for GeoSurv II Fuselage 60
5.2.1. Structural Foam Core Materials 61
5.2.2. Balsa Wood Cores 65
5.2.3. Other Core Materials 66
5.2.4. Core Selection for GeoSurv II Fuselage 67
5.3. Selection of Matrix and Reinforcement Materials 69
5.4. Material Selection for Rigid Inserts 70
CONTENTS vT
![Page 8: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/8.jpg)
CHAPTER 6. FUSELAGE REDESIGN 72
6.1. Redesign objectives 72
6.2. Design Changes to the GeoSurv II Fuselage 73
CHAPTER 7. DESIGN OPTIMIZATION 80
7.1. GeoSurv II Fuselage FEA 80
7.1.1. FE Model Construction 80
7.1.2. Material Properties 82
7.1.3. Part Meshing Considerations 86
7.1.4. FE Model Assembly and Constraints 89
7.1.5. Analysis Steps, Loads and Boundary Conditions 90
7.1.6. Mesh Independence of the Results 93
7.1.7. FEA Simulations 95
7.1.8. FEA Results 97
7.2. Experimental Verification of the FEA results 102
7.2.1. Test Matrix, Specimen Manufacturing and Test Procedure 103
7.2.2. FEA Simulations 106
7.2.3. Results 108
CHAPTER 8. FUSELAGE MANUFACTURING 113
8.1. Sample Section Manufacturing 113
8.2. CCBM Experiments and Fuselage Manufacturing Model 117
8.3. Fuselage Manufacturing 121
CHAPTER 9. MANUFACTURING RESULTS 128
9.1. Surface Finish, Weight and Tolerances 128
9.2. Process Viability 131
CHAPTER 10. CONCLUSIONS 133
10.1. Conclusions 133
10.2. Recommendations for Future Work 134
REFERENCES 137
APPENDICES 141
Appendix A: Manufacturing Supplies 141
Appendix B: CCBM Bag Manufacturing Procedure 142
Appendix C: Process Value Analysis 145
Appendix D: Core Materials and Inserts 148
CONTENTS vii
![Page 9: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/9.jpg)
Appendix E: FEA Results and Weight Estimates 157
Appendix F: Microscopic Image Analysis 158
Appendix G: Fuselage Profiling 160
Appendix H: Fibre Volume Fraction Calculation 168
CONTENTS viii
![Page 10: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/10.jpg)
LIST OF TABLES
Table 2.1: Airframe requirements for the GeoSurv II UAS [5] 15
Table 2.2: Design and manufacturing objectives for the new fuselage 15
Table 3.1: Potential advantages of disadvantages of LCM processes [6-10] 18
Table 3.2: New materials and technologies developed to improve VARTM process 25
Table 3.3: Advantages and Disadvantages of CCBM process [24] 44
Table 4.1: PVA matrix: processes for mouldless CCBM/VARTM 55
Table 5.1: Manufacturing requirements for the core material 61
Table 5.2: Structural foam cores suitable for mouldless VARTM manufacturing 63
Table 5.3: Normalized mechanical properties of the most structural foam cores 68
Table 5.4: Carbon fibre fabric specifications 69
Table 6.1: GeoSurv II fuselage redesign: goals and limitations 72
Table 7.1: Properties of Airex C PVC foam 83
Table 7.2: Properties of the current fuselage materials 84
Table 7.3: Derivation of lamina properties for the new fuselage 85
Table 7.4: Properties of pins and bushings 86
Table 7.5: Parametric Study Results 94
Table 7.6: FEA verification test matrix 103
Table 7.7: Bearing test failure loads 108
Table 7.8: Bearing test results: FEA predictions and Experiments 109
Table 9.1: Comparison of major dimensions: fuselage design vs. current and new fuselages.. 130
Table 9.2: Comparison of the actual weight and predicted weight of the new fuselage 131
LIST OF TABLES ix
![Page 11: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/11.jpg)
LIST OF FIGURES
Figure 1.1: Current configuration and specifications of GeoSurv II UAS (2009/2010) [4] 5
Figure 1.2: GeoSurv II Prototype Assembly (February 2010) 5
Figure 2.1: Main sections of the GeoSurv II fuselage [3] 10
Figure 2.2: Cross-section of the fuselage structural H frame [3] 11
Figure 2.3: Sequence of the main steps employed in mouldless VARTM 12
Figure 2.4: Fuselage main frame manufactured by mouldless VARTM 14
Figure 2.5: Fuselage prepared for assembly 14
Figure 2.6: Work-flow diagram of the new fuselage development 16
Figure 3.1: A family of state of the art LCM processes 20
Figure 3.2: Schematic of the Basic RTM setup 21
Figure 3.3: SSRIM process [8] 22
Figure 3.4: Schematic of basic VARTM process 23
Figure 3.5: Schematic of the EADS VAP® before (top) and after (bottom) infusion [10] 26
Figure 3.6: Schematic of the FASTRAC process [18] 27
Figure 3.7: Schematic of the CAPRI process [19] 28
Figure 3.8: Schematic of double bag VARTM 29
Figure 3.9: Outline of NCW fabric [20] 30
Figure 3.10: Sequence of steps in A-VARTM process [20] 31
Figure 3.11: Illustration of SLI process [21] 31
Figure 3.12: Pressure distribution during and after resin injection in SLI process [21] 32
Figure 3.13: Typical CCBM procedure 33
Figure 3.14: Schematic of the CIRTM method [25] 35
Figure 3.15: Details of the part during layup [26] 36
Figure 3.16: Schematic of the EC-HLM process [26] 36
Figure 3.17: Schematic of the RIDFT process [10] 39
Figure 3.18: Industrial RIDFT machine 10 ft x 15 ft x 4 ft [29] 39
Figure 3.19: Flexible Injection Process [30] 41
Figure 3.20: Schematic of the RFI process [31] 42
Figure 3.21: A part bagged with disposable vacuum bag [32] 45
LIST OF FIGURES x
![Page 12: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/12.jpg)
:igure 4.1: Common reusable seal configuration for CCBM 47 :igure 4.2: Components of Arctek CCBM system [35] 49 :igure 4.3: CCBM bag manufacturing trial #1: tool setup 50 :igure 4.4: CCBM bag manufacturing: trial #1 results 50 :igure 4.5: CCBM bag manufacturing trial #2: tool setup 51 :igure 4.6: CCBM bag manufacturing: trial #2 results 52 :igure 4.7: Sample CCBM bag section with resin distribution channels 52 :igure 4.8: Flow profiles of CIB method (a) and disposable distribution media (b) 53 :igure 4.9: Cost of the fuselage for increasing part count at labour rate of 20$/hr 56 :igure 4.10: Cost of the fuselage for increasing part count at labour rate of 40$/hr 57 :igure 5.1: Sandwich beam subjected to three point bend 58 :igure 5.2: Sandwich beam in bending 59 :igure 5.3: Balsa wood core-end grain configuration [43] 65 :igure 5.4: Comparison of the selected foam materials at 4 lbs/ft3 density 68 :igure 5.5: Comparison of inserts for sandwich assembly 71 :igure 6.1: Fuselage redesign work-flow diagram 73 :igure 6.2: Current and the new fuselage wall design (units: in.) 74 :igure 6.3: Current and the new bolted sandwich assembly 75 :igure 6.4: New landing gear attachment plate 76 :igure 6.5: Current and the new landing gear configurations 76 :igure 6.6: Fuselage wall straight section extension 77 :igure 6.7: Design modification at the fairings 77
igure 6.8: Increased core thickness at the locations of the fasteners 78
igure 6.9: Core extension to mount the nosecone 78
igure 6.10: Reinforcement for flight avionics rack 78
igure 6.11: Current and the Modified fuselage concept models 79
igure 7.1: Parts modelled for the fuselage FEA 81
igure 7.2: Partitions created on the fuselage skin for meshing 88
igure 7.3: Mesh refinement near the pin holes 88
igure 7.4: Fuselage FE model assembly 89
igure 7.5: Tie constraints established between the pins and the fuselage structure 90
LIST OF FIGURES xi
![Page 13: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/13.jpg)
:igure 7.6: Boundary Conditions 91 :igure 7.7: Wing lift and engine loads 91 :igure 7.8: Main landing gear loads 92 :igure 7.9: Mission avionics and nose landing gear loads 92 :igure 7.10: Substructure model used for parametric study 93 :igure 7.11: Mesh refinement: parametric study level 1 to level 12 94 :igure 7.12: Von Mises stress convergence (5%) 95 :igure 7.13: Fuselage FE model 96 :igure 7.14: Optimized foam core for the new fuselage 97 :igure 7.15: Optimized skin layup for the new fuselage 98 :igure 7.16: Von Mises (left) and in-plane shear (right) stresses (psi) in the skin under flight oads 100 :igure 7.17: Von Mises stresses (psi) in the skin during the landing step 101 :igure 7.18: Shear stresses (psi) in the skin under landing loads 101 :igure 7.19: Loading modes chosen for experiments 102 :igure 7.20: Geometry of the sandwich specimen 104 :igure 7.21: Bearing test setup in the load frame 105 :igure 7.22: Bending test setup in the load frame 106 :igure 7.23: Abaqus model showing the loads and boundary condition of the specimens 107 :igure 7.24: Abaqus FE model of the test specimens 107 :igure 7.25: Bearing test: force-displacement data 109 :igure 7.26: Comparison of the failure loads in bearing test 110 :igure 7.27: Close-up of the bearing failure mode 110 :igure 7.28: Bending test FEA prediction 112
igure 7.29: Bending test: force-displacement data 112
igure 8.1: Geometry of the test article 115
igure 8.2: Important features of mouldless CCBM setup 116
igure 8.3: Manufactured component 116
igure 8.4: Bondline comparison of VARTM and CCBM manufactured sandwich coupons 117
igure 8.5: CCBM experiment setup 118
igure 8.6: Conceptual CCBM Manufacturing Model 120
LIST OF FIGURES xii
![Page 14: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/14.jpg)
Figure 8.7: Mouldless CCBM Process 121
Figure 8.8: Machining of the foam parts on the CNC router table 122
Figure 8.9: Foam parts required for fuselage manufacturing 123
Figure 8.10: FRP inserts for fuselage 123
Figure 8.11: Bonding of foam parts in the fixture 124
Figure 8.12: Features included in the foam parts to facilitate precise assembly 124
Figure 8.13: Mouldless CCBM setup 126
Figure 8.14: Resin starved regions observed during the infusion 127
Figure 9.1: Fuselage Manufactured by mouldless CCBM 128
Figure 9.2: Revised PVA cost estimates based on actual labour required for fuselage manufacturing (labour rate $40/hr) 132
LIST OF FIGURES xiii
![Page 15: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/15.jpg)
CHAPTER 1. INTRODUCTION
This chapter provides an overview of the ongoing low-cost composite airframe research at
Carleton University. The structural demonstrator used in this research, the GeoSurv II
Unmanned Aircraft System (UAS) is also discussed. The thesis objectives are outlined, followed
by the procedure utilized to accomplish the objectives and the thesis contributions.
1.1. Background
Sandwich construction has been used in primary and secondary aircraft structures for many
years. It consists of thin face-sheets or skins adhesively bonded to both surfaces of a relatively
thick, low density core material. This leads to increased strength and stiffness for little added
weight. A properly designed sandwich construction also offers many other advantages such as,
thermal insulation, impact resistance and noise attenuation [1,2].
For many decades, aluminum and Nomex® honeycomb have been the two most commonly
used core materials in aerospace applications; they offer excellent specific strength-to-weight
and stiffness-to-weight ratios. However, their open and anisotropic cell structure leads to
several problems such as core crushing during shaping, curing or when subjected to lateral
forces. Honeycomb based sandwich components also exhibit extensive moisture ingress and
consequent corrosion, which lead to component service restrictions as well as increased life
cycle costs. Additionally, processing of honeycomb sandwich components typically requires
expensive autoclaves and the use of pre-impregnated composite fabrics (prepregs) [2].
Various foam and balsa core technologies have successfully addressed some of these issues at
much lower costs, with their closed cells and chemically resistant constructions. The
CHAPTER 1. INTRODUCTION 1
![Page 16: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/16.jpg)
development of out-of-autoclave Liquid Composite Moulding (LCM) processes further enhances
the cost savings, making sandwich construction an economical solution for many structural
applications [3,4].
Developing low cost manufacturing methods without sacrificing part quality, repeatability and
performance, is always in the best interest of aerospace companies, specifically those dealing
with small aircraft and UASs for civilian applications. Costs for small aerospace companies can
be significantly reduced with the use of low cost manufacturing methods. This was the main
driving factor for the inception of low-cost composite airframe research at Carleton University.
1.2. Overview of the Low-Cost Composite Airframe Research
Optimizing the design and manufacturing methods to reduce cost has been an ongoing area of
research at Carleton University for the past four years. The research addresses one major
barrier to widespread use of composite materials in aircraft structures: high material and
manufacturing costs. Conventional prepreg layup and autoclave curing methods produce
composite parts with optimum fibre volume fraction, low void content and excellent surface
finish. However, the prepregs and autoclave infrastructure come with high initial and recurring
costs, which are beyond the means of small companies. Hence, the aerospace industry is
constantly searching for low cost, out-of-autoclave manufacturing methods capable of
producing high quality components. Such manufacturing methods will directly benefit small
companies, such as those that deal with general aviation aircraft and UAS [3,4].
The early stages of this research were funded by the Ontario Government, through the Ontario
Centre of Excellence (OCE) Interact program and Sander Geophysics Ltd. (SGL). Subsequent
CHAPTER 1. INTRODUCTION 2
![Page 17: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/17.jpg)
financial support was also provided by SGL, along with an NSERC Collaborative Research and
Development (CRD) grant and support from the National Research Council - Institute of
Aerospace Research (NRC-IAR).
The low-cost composite airframe research uses the GeoSurv II UAS as a technology
demonstrator, to develop and implement the processes that are capable of producing
structural components of varying complexity. GeoSurv II is an excellent demonstrator, due to its
relatively small size and lower reliability requirements compared to a manned aircraft.
During the academic years 2006-2008, low cost processing of carbon-epoxy composite
components was investigated by Maley [3] and Vacuum Assisted Resin Transfer Moulding
(VARTM) was identified to be a suitable process for manufacturing most of the GeoSurv II UAS
components. Maley also suggested that manufacturing of sandwich structure components in
small quantities could be achieved economically through a single step "mouldless" infusion.
This process uses the core material as the mould to fabricate composite sandwich structures,
eliminating the material and labour costs associated with mould preparation. Hence, this
process can be very beneficial to aerospace companies for producing small aircraft and UASs in
low production quantities or for rapid prototyping of geometrically complex sandwich
components.
In order to prove the viability of the process, Maley implemented mouldless VARTM on the full
scale fuselage section of the GeoSurv II UAS [3]. The outcome showed a need for improvement
in the process robustness, repeatability, and tolerances. The following sections provide a brief
CHAPTER 1. INTRODUCTION 3
![Page 18: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/18.jpg)
overview of the GeoSurv II UAS followed by the summary of thesis objectives, thesis
organization and contributions.
1.3. Overview of the GeoSurv II UAS
GeoSurv II is an all-composite UAS, currently being developed at Carleton University as part of a
fourth year undergraduate team project in the Department of Mechanical and Aerospace
Engineering. It is designed to perform multi-purpose geophysical survey missions including high
resolution magnetic surveys that are of particular interest to the industrial partner SGL, an
Ottawa based company specializing in airborne geomagnetic, gravimetric and radiometric
surveys around the world.
SGL currently flies airborne surveys using conventional manned fixed-wing aircraft and
helicopters. A survey mission requires a minimum crew of four people: two pilots, a
geophysicist and an aircraft maintenance engineer. The GeoSurv II is designed to be controlled
autonomously with an autopilot featuring an on-board obstacle detection and avoidance
system. This will enable geophysical surveys to be executed with two operators: a geophysicist
and an aircraft maintenance engineer. The autonomous nature of the GeoSurv II will, in future,
allow the execution of multiple aircraft operations using a single operator, thereby increasing
the quantity of the data collected at reduced operational costs.
The composite airframe design of the GeoSurv II reduces undesirable magnetic noise within
near proximity of the magnetometers mounted on the aircraft wing tips. Benefits of composite
airframe construction coupled with the GeoSurv ll's ability to fly slower and closer to the
ground will substantially improve the quality of the acquired data at much lower capital and
CHAPTER 1. INTRODUCTION 4
![Page 19: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/19.jpg)
operational costs compared to its manned counterparts. Figure 1.1 shows the current
configuration and specifications of the GeoSurv II [3].
v*& •Wing span 16 ft. 11 Length 14 ft. 11 Height 3 ft. •200lbMTOW •60/100 kts cruise speed •30 hp propeller engine with pusher configuration •Composite airframe
Figure 1.1: Current configuration and specifications of GeoSurv II UAS (2009/2010) [4]
The 2010/2011 academic year is the 7th year of GeoSurv II development at Carleton University.
Thus far, most of the design and analysis of GeoSurv II has been completed and a working
prototype has been constructed. The prototype (Figure 1.2) is currently being prepared for
initial flight testing.
Figure 1.2: GeoSurv II Prototype Assembly (February 2010)
Low-cost composite airframe research is one of five critical areas, which required more
advanced graduate level research during the course of the GeoSurv II development. Other
graduate research topics include autonomous operation, obstacle detection and avoidance,
flight control actuating systems with low magnetic signature and geomagnetic data acquisition.
CHAPTER 1. INTRODUCTION 5
![Page 20: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/20.jpg)
1.4. Thesis Objectives and Organization
This thesis work focuses on the design and manufacturing of geometrically complex, foam-core
composite sandwich structures, by means of "mouldless" LCM method. The thesis objective is
to develop a robust, low-cost mouldless LCM method for foam-core composite sandwich
structures and utilize the method to manufacture a full scale GeoSurv II fuselage.
In this work, the design of GeoSurv II fuselage and previously implemented mouldless VARTM
methods were studied to identify critical areas that require improvement. Based on this study,
design and manufacturing objectives were proposed for the next generation fuselage. Then,
state-of-the-art LCM processes found in literature were reviewed, which identified Closed
Cavity Bag Moulding (CCBM) as a potential alternative to VARTM for mouldless manufacturing.
A series of experiments was carried out using CCBM process to evaluate its feasibility for
mouldless manufacturing, including investigations of various bag sealing and resin distribution
methods. Feasibility of CCBM for mouldless manufacturing was assessed through a Process
Value Analysis (PVA). Appropriate materials were selected for fuselage manufacturing through
the mouldless CCBM method. Subsequently, the fuselage was redesigned for improved
manufacturability. The new design was optimized using a Finite Element Analysis (FEA)
implemented in Abaqus. The FEA predictions were verified by testing demonstrator coupons at
specific loading conditions. Finally, a full scale GeoSurv II fuselage was manufactured to
demonstrate the viability of the developed process.
CHAPTER 1. INTRODUCTION 6
![Page 21: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/21.jpg)
This work is organized into the following chapters:
> Chapter 2: Overview of the GeoSurv II Fuselage The design of the GeoSurv II fuselage is discussed. A mouldless VARTM method previously implemented on the fuselage prototype is presented. A post-process assessment is also provided in conjunction with the design and manufacturing objectives for the new fuselage.
> Chapter 3: Review of LCM Processes A summary of the available LCM technologies is presented. Various low cost LCM processes are assessed to select the most suitable process for mouldless manufacturing.
> Chapter 4: Mouldless CCBM Process Development A series of experiments carried out using the chosen CCBM method from Chapter 1 are discussed. Various bag sealing methods and infusion strategies are evaluated to determine the feasibility of CCBM for mouldless manufacturing. Finally, PVA carried out to determine the most feasible mouldless manufacturing method is presented.
> Chapter 5: Fuselage Material Selection State-of-the-art structural sandwich core materials are discussed and suitable core material for mouldless manufacturing is selected. Material choices for the matrix, reinforcement and inserts are also presented along with the rationale for selection.
> Chapter 6: Fuselage Redesign The design changes made to the GeoSurv II fuselage to improve its manufacturability are described. An improved GeoSurv II fuselage model is presented.
> Chapter 7: Design Optimization An FEA carried out on the new fuselage design, to optimize the composite layup is discussed. A weight estimate for the optimized structure is provided. Finally, an experimental verification of the FEA results using demonstrator test coupons is presented.
> Chapter 8: Fuselgge Manufacturing Manufacturing of a full scale GeoSurv II fuselage by an improved mouldless CCBM method is discussed.
> Chapter 9: Manufacturing Results The outcomes of the fuselage manufacturing by mouldless CCBM method are discussed. The results are assessed against the design and manufacturing goals.
> Chapter 10: Conclusions Conclusions drawn from this research work are summarized. Recommendations and future work are discussed.
CHAPTER 1. INTRODUCTION 7
![Page 22: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/22.jpg)
1.5. Contributions
In this research, state-of-the-art LCM processes currently used for manufacturing composite
components were studied and the processes applicable for mouldless manufacturing were
identified. An in-depth assessment of the applicable processes showed that CCBM should be
considered further for mouldless manufacturing.
CCBM is a relatively new process, currently popular in the marine industry. This resin infusion
process uses silicone based elastomeric vacuum bags that are form-fitted to the shape of the
part. CCBM infusion offers good vacuum integrity and better surface finish compared to
traditional VARTM infusion using disposable vacuum bags. However, current CCBM methods
are intended to be used for manufacturing with rigid moulds and hence can be expensive at the
outset [4]. As part of this research, a series of experiments were carried out to develop CCBM
techniques suitable for mouldless manufacturing. A PVA was used to assess the process
variants and select the most suitable technique (section 4.3) for mouldless manufacturing. This
research laid the foundation for further development of mouldless CCBM methods.
The use of Design for Manufacturing (DFM) principles and FEA for continuous design
improvement was demonstrated. An effective methodology for creating bolted joints in foam
core composite sandwich structures was developed and demonstrated.
DFM principles were applied to improve the manufacturability of the GeoSurv II fuselage. The
new design was optimized using a simplified FEA carried out using Abaqus. An experimental
study was conducted to verify selected FEA results. Approximately 36% weight savings were
CHAPTER 1. INTRODUCTION 8
![Page 23: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/23.jpg)
achieved with this design optimization work. The Abaqus FE model of the fuselage can later be
expanded to include more realistic material property formulations and dynamic loading.
This thesis has contributed to the development of a unique mouldless manufacturing technique
for foam-core composite components, using an improved CCBM method featuring Channel-ln-
Bag (CIB) infusion. A full scale GeoSurv II fuselage has been manufactured in a single infusion
step, using this mouldless manufacturing method. This is a low cost LCM process suitable for
producing complex geometry foam-core composite components with good part quality and
process repeatability. The process is economical for part quantities below 10 and hence is
beneficial for low to medium volume production runs and prototype development. This
manufacturing method should be considered for small aircraft and UASs featuring foam-core
composite designs.
This research has successfully addressed the major aspects of low cost composite structures
including structural design with material selection, structural optimization, DFM, PVA,
manufacturing process planning and development.
CHAPTER 1. INTRODUCTION 9
![Page 24: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/24.jpg)
CHAPTER 2. OVERVIEW OF THE GEOSURV II FUSELAGE
This chapter provides an overview of the GeoSurv II UAS fuselage and describes the mouldless
VARTM method previously implemented for manufacturing the first fuselage prototype. A post
process assessment is also provided followed by the design and manufacturing objectives for
the new fuselage.
2.1. Design of the GeoSurv II fuselage
Major components of the current GeoSurv II fuselage assembly are shown in Figure 2.1. It
consists of a structural main frame ("H-frame"), upper and lower access hatches, and a
nosecone. The access hatches and the nosecone contribute mainly to aerodynamic drag
reduction and are subjected to relatively low structural loads, with the exception of the lower
access hatch, which carries 20 lbs of geomagnetic survey payload.
Figure 2.1: Main sections of the GeoSurv II fuselage [3]
CHAPTER 2. OVERVIEW OF THE GEOSURV II FUSELAGE 10
![Page 25: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/25.jpg)
The structural main frame features an "H" cross-section with foam core carbon-epoxy sandwich
construction. Figure 2.2 shows the cross-section of the main frame and highlights critical areas
through which the in-flight and the landing loads are transferred into the fuselage. The in-flight
bending stresses from the wings are transferred through the carry-through spar, while the
aerodynamic moments are transferred through two shear pins located forward and aft of the
spar.
The front and rear bulkheads carry the nose landing gear loads and the engine loads
respectively. Two bolts directly above the carry-through spar location transfer loads from the
main landing gear into the fuselage walls. An airfoil shaped fairing is incorporated in the design
to provide smooth transition from the fuselage to the wings and thus minimize interference
drag.
Rear bulkhead: Engine loads
Main iandinggearattachment Aft shear pin location
Carry through sparlocation
Forward shear pin location
Front bulkhead: Nose Landing gear/Air data boom loads
Figure 2.2: Cross-section of the fuselage structural H frame [3]
CHAPTER 2. OVERVIEW OF THE GEOSURV II FUSELAGE 11
![Page 26: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/26.jpg)
2.2. Manufacturing of the Fuselage Prototype
A prototype of the fuselage main frame was manufactured using a mouldless VARTM method
developed by Maley [3]. Mouldless VARTM is a low cost processing method applicable for
composite sandwich components, in which the foam core is used as the mould during
manufacturing. This manufacturing method includes three primary steps: core preparation,
fabric layup and infusion, all of which are shown Figure 2.3. In step 1 "Core Preparation", the
required components were machined from Extruded Poly Styrene (EPS) insulation foam1. These
foam components were assembled using Airtac 2 spray adhesive. The foam assembly was then
coated with a thin layer of West System 105 epoxy resin and finished with West System 407 (a
low density fairing filler), both of which were used to increase the core stiffness. In step 2:
"Fabric Layup", the prepared foam component (the core assembly) was laid-up with Hexcel -
AGP 370 5 HS fabric at [(0/90)°, (+/- 45)°, Foam Core]SYM orientation.
Figure 2.3: Sequence of the main steps employed in mouldless VARTM
EPS (pink) foam was used in the GeoSurv II prototype, primarily to reduce cost. Using an aerospace grade structural core material would aid in better dimensional tolerances and may result in significant weight saving, as it eliminates the need to seal the foam prior to infusion.
CHAPTER 2. OVERVIEW OF THE GEOSURV II FUSELAGE 12
![Page 27: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/27.jpg)
In step 3: "Infusion", optimum locations for resin inlets and outlets were determined through
the use of Liquid Injection Moulding Simulation (LIMS) software [3]. In this process, 1-D
permeability values were first measured for the fuselage layup. These permeability values were
used with the LIMS software to simulate the resin flow during infusion. Simulation results were
used to determine the optimal locations of both resin inlets and outlets, to improve infusion
quality and to reduce infusion time. The infusion setup was prepared with the resin inlet and
outlet lines placed at the locations suggested by the flow simulation. The fuselage was then
manufactured in a single infusion step. A detailed description of this process simulation and
manufacturing method can be found in [3].
2.3. Post Process Assessment of the GeoSurv II Fuselage
The prototype fuselage manufactured by mouldless VARTM, is shown in Figure 2.4. One of the
major problems encountered was the occurrence of vacuum bag leaks during resin infusion.
The leaks resulted in small air pockets at the corners of the cured fuselage. Furthermore, the
foam core was distorted under the applied vacuum pressure, leading to an average deviation
from flatness of 0.15 in. over the side wall (length: 44 in.) and up to 0.31 in. deviation from the
target dimension at several locations [3]. The manufactured component was sanded and filled
to reduce the dimensional variations and improve the surface finish. The access hatches and
several other components that interface with the fuselage main frame were modified to mate
with the distorted shape of the as-manufactured fuselage. Additionally, extensive labour was
required to complete the fuselage assembly, shown in Figure 2.5. This included drilling holes
and bonding rigid inserts into the holes to facilitate load transfer into the structure at various
joints, manufacturing and bonding of the nose cone bridges, bay separator panel and mounting
CHAPTER 2. OVERVIEW OF THE GEOSURV II FUSELAGE 13
![Page 28: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/28.jpg)
brackets for the mission avionics rack. Thus, the cost of labour required to carry the fuselage
into a finished stage outweighed the economic advantages of mouldless VARTM. This indicated
a need for improvement to both the structural design and the manufacturing process.
Figure 2.4: Fuselage main frame manufactured by mouldless VARTM
Upper nosecone bridge
Holes/inserts for mountingNLG and air data boom
Wooden brackets to mountthe nosecone
Figure 2.5: Fuselage prepared for assembly
CHAPTER 2. OVERVIEW OF THE GEOSURV II FUSELAGE 14
![Page 29: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/29.jpg)
2.4. Design and Manufacturing Objectives for the New Fuselage
Enhanced with the appropriate selection of new materials and an improved processing method,
the new fuselage shall replace the prototype in its present configuration. The new fuselage shall
be manufactured using a low-cost, mouldless LCM method. The manufactured component shall
comply with the GeoSurv II airframe requirements, listed in Table 2.1. The design and
manufacturing objectives of the new fuselage are given in Table 2.2. The work-flow involved in
the development of the new fuselage is shown in Figure 2.6.
Table 2.1: Airframe requirements for the GeoSurv II UAS [5]
GeoSurv II UAS Airframe Requirements > Materials used in the primary and secondary structures shall be non-ferrous to the greatest extent
possible, to minimize magnetic noise. > Light weight modular airframe. > The structure shall be robust and relatively inexpensive to manufacture and maintain. > The UAS should operate within a temperature range of -49 to 104 °F (-45 to 40 °C) > The UAS airframe should comply with Canadian Aviation Regulation, Part V, 523-VLA (Very Light
Aircraft): Fire/smoke resistance, moisture, chemical corrosion resistance.
Table 2.2: Design and manufacturing objectives for the new fuselage
Re-design Objectives > Improve the design to reduce weight and
increase structural integrity. > Apply effective Design for Manufacturing
(DFM) and Integrated Manufacturing (IM) principles to facilitate near-net-shape manufacturing by mouldless LCM method.
Manufacturing Objectives > Improve part quality and tolerances (near-net-
shape manufacturing). > Improve process repeatability.
Note: The first prototype of the GeoSurv II fuselage developed by Maley [3] is referred to as the
current fuselage, and the fuselage developed as part of this research is referred to as the new
fuselage throughout the rest of this thesis.
CHAPTER 2. OVERVIEW OF THE GEOSURV II FUSELAGE 15
![Page 30: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/30.jpg)
Problem Definition Review of the current fuselage
design/manufacturing (Chapter2)
Literature Review and Process Selection
(Chapter 3)
Process Optimization and Feasibility Studies
(Chapter4)
Material Selection (Chapters)
Design Optimization (Chapter 7)
New Fuselage Design (Chapter6)
New Fuselage Manufacturing (Chapters: 8,9)
Figure 2.6: Work-flow diagram of the new fuselage development
CHAPTER 2. OVERVIEW OF THE GEOSURV II FUSELAGE 16
![Page 31: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/31.jpg)
CHAPTER 3. REVIEW OF LCM PROCESSES
This chapter provides an introduction to LCM processes and discusses several LCM variants
used in the industry, for manufacturing composite components. Following the review, low cost
LCM processes applicable for mouldless manufacturing were assessed to identify the most
suitable process.
3.1. LCM Processes
LCM refers to a family of processes, in which dry fibrous reinforcement materials placed in a
mould cavity are impregnated with a liquid polymer matrix (typically thermoset resins) under a
forcing pressure gradient (vacuum pressure, positive pressure or both). The resin is then
allowed to cure; the part is demoulded and subjected to finishing operations as required, to
create the final product. LCM processes can be implemented using many low-cost tooling and
part constituents for a wide range of part size, complexity and production quantity.
Characterized by a closed mould setup, LCM processes further offer an operator-friendly
manufacturing environment, good fibre orientation control and improved process repeatability.
Along with these benefits there exist some inevitable drawbacks, such as relatively complex
processing steps and difficult quality control with in-house resin mixing [6-9]. The potential
advantages and disadvantages of LCM processes compared to other traditional composite
manufacturing methods are summarized in Table 3.1.
CHAPTER 3. REVIEW OF LCM PROCESSES 17
![Page 32: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/32.jpg)
Table 3.1: Potential advantages of disadvantages of LCM processes [6-10]
Advantages > Low material, initial equipment and recurring
costs compared to prepreg-autoclave method > Many material choices for reinforcement,
matrix and core > No size restrictions: Integrated manufacturing
of large structural components is possible > Low cost LCM methods yield more consistent
part quality than wet layup and vacuum bagging, at similar costs
> Greater flexibility than any other processing method
> Thick laminates can be processed easier than with wet layup
> Produced laminates have uniform microstructure and minimum void content compared to wet layup
> Sandwich constructions can be produced in single infusion process
> Characterized by closed-mould operations, the LCM methods reduce volatile organic compounds (VOC) emission from resin by more than 90% as compared to open-mould hand lamination methods
Disadvantages > Requirement for relatively complex/different
skills than wet layup > Requirement for resins with low viscosity may
compromise the thermal and mechanical properties of some polyester/vinyl based laminates
> Low fibre volume fraction compared to prepreg-autoclave method (Except for RTM type processes)
> Air leaks in the tool or bag and uneven resin flow may result in resin rich/starved regions and in turn cause expensive scrap parts
> With the exception of RTM type processes and LRTM, most other LCM processes produce parts with moulded finish on one side.
> Difficult to implement on the honeycomb cores
Attracted by the potential cost savings and process flexibility, manufacturers are increasingly
researching and developing LCM techniques for producing their components. This has led to
the evolution of many LCM process variants over the last few decades. Despite the complexities
associated with specific process variants, all LCM processes share several distinctive features
[7,8]:
> A resin delivery system > A mould setup equipped with appropriate clamping and manipulation devices > A reinforcement handling system (i.e. fibre preforms, sheet/bulk mounding compounds) > A strategy for air displacement or evacuation and resin supply
CHAPTER 3. REVIEW OF LCM PROCESSES 18
![Page 33: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/33.jpg)
The functionality of these systems is determined by the scale of the part and complexity of the
manufacturing operation.
Based on the forcing pressure gradient used to introduce the resin into the mould cavity, the
LCM process variants can be grouped into two categories: Resin Transfer Moulding (RTM) and
Vacuum Assisted Resin Transfer Moulding (VARTM). The former uses positive pressure to inject
the resin into the mould cavity, while the latter uses a negative pressure gradient to draw the
resin into the mould cavity. Figure 3.1 shows a family of state of the art LCM processes
developed from RTM and VARTM. Unique features of each of these processes are described in
the following sections.
CHAPTER 3. REVIEW OF LCM PROCESSES 19
![Page 34: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/34.jpg)
Liquid Composite
Moulding (LCM) Processes
Resin Transfer Moulding
(RTM) & Structural
Reaction Injection
Moulding (SRIM)
Vacuum Assisted Resin
Transfer Moulding
(VARTM)/VAP/VIMP
- »
Seemann Composites Resin Transfer Moulding (SCRIMP™)
Vacuum Assisted Process (VAP®)
Fast Remotely Actuated Channelling (FASTRAC)
Controlled Atmospheric Pressure Resin Infusion (CAPRI)
Double Bag VARTM
Advanced VARTM (A-VARTM)
Single Line Injection (SLI)
High Performance VARTM/RTM (Hyper-VARTM™, Hyper-RTM™)
Vibration Assisted Liquid Composite Moulding
Closed Cavity Bag Moulding (CCBM)
Co-Injection Resin Transfer Moulding (CIRTM)
Euro Composites® Honeycomb Liquid Moulding (EC-HLM)
VacFlo® Process
Light RTM
Resin Infusion between Double Flexible Tooling (RIDFT)
Flexible Injection
Resin Film Infusion (RFI)
Semi-Preg Infusion
Figure 3.1: A family of state of the art LCM processes
3.1.1. Resin Transfer Moulding (RTM)
RTM is best known for its ability to fabricate large, complex parts to near-net shape with
excellent surface finish on both sides. A schematic of a basic RTM setup is shown in Figure 3.2.
In this process, dry fibre preform is laid-up into a two part rigid mould, and resin is injected into
the mould cavity until the preform is fully saturated. The part is then cured and removed from
the mould. RTM produces near autoclave quality parts with void content <1%, at very short
cycle times (typically 3-5 minutes). However, it requires matched, leak proof moulds capable of
CHAPTER 3. REVIEW OF LCM PROCESSES 20
![Page 35: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/35.jpg)
withstanding the injection pressures up to 1500 psi. Designing and manufacturing RTM moulds
require high initial costs, making RTM economical for large production quantities, typically
above 1000 units [10,11].
Resin Injection Port
ClampingPressure
Figure 3.2: Schematic of the Basic RTM setup
3.1.2. Structural Reaction Injection Moulding (SRIM)
SRIM is an LCM process evolved from the family of Injection Moulding (IM) processes. IM is a
process whereby liquid thermoplastic or partially polymerized thermoset resins are injected
into mould cavity to produce non-reinforced polymeric components. This process is typically
carried out at elevated temperatures and pressures for rapid part production in large
quantities. A further development to IM is the Reaction Injection Moulding (RIM): a process in
which two part thermoset resins are mixed and injected into the mould cavity to produce the
final part. The processes take their name from the chemical reaction that takes place within the
machine [8]. When short fibre reinforcements are included into the RIM, the process is called
Reinforced Reaction Injection Moulding (RRIM). SRIM is a further development to RRIM,
whereby preformed fibre mat reinforcements are impregnated with liquid resin to produce the
CHAPTER 3. REVIEW OF LCM PROCESSES 21
![Page 36: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/36.jpg)
final part. The sequence of steps employed in SRIM is shown in Figure 3.3. This process is
characterized by mould fill times less than 1 minute and pressures greater than 1500 psi.
Figure 3.3: SSRIM process [8]
The RTM and SRIM equipments, though relatively inexpensive compared to an autoclave
infrastructure, require somewhat high initial costs. Hence, they are efficient for high volume
production runs. Since the primary focus of this research is low volume production by LCM
process, RTM and its process variants were not studied in detail. Substantial reviews of RTM
type processes can be found in references [8 -11].
3.1.3. Vacuum Assisted Resin Transfer Moulding (VARTM)
The need for an LCM process for economic part production in low production quantities has
resulted in the development of Vacuum Assisted Resin Transfer Moulding (VARTM). VARTM is a
variation of RTM, in which resin is drawn into the preform under vacuum pressure, rather than
being injected at positive pressure. A schematic of the VARTM process is shown in Figure 3.4.
CHAPTER 3. REVIEW OF LCM PROCESSES 22
![Page 37: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/37.jpg)
This process begins with preparation of a rigid mould and layup of dry reinforcement in the
mould cavity. Then, a flexible vacuum bag is placed over the reinforcement and is sealed
against the mould using sealant tape. A vacuum tight chamber is created inside the bag by
connecting the outlet to a vacuum pump. Resin is then introduced into the mould via an inlet,
which is typically on the opposite side of the outlet. The reinforcement is impregnated by the
resin due to the pressure differential between the inlet and the outlet ports. When the
reinforcement is fully impregnated, the resin is left to cure under the vacuum pressure to
create the final part. In order to reduce the flow resistance and hence the filling time, a highly
permeable distribution medium is introduced between the top layer of the reinforcement and
the vacuum bag. During infusion, the resin distribution medium lifts the bag surface away from
the preform, leaving a highly permeable path for resin to travel. Resin is distributed
preferentially through the surface of the flow media and simultaneously through thickness of
the preform [3,12,13].
Figure 3.4: Schematic of basic VARTM process
VARTM is a low-cost process capable of manufacturing large-scale complex geometry
composite parts with good surface finish and excellent dimensional tolerances. This processing
CHAPTER 3. REVIEW OF LCM PROCESSES 23
![Page 38: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/38.jpg)
method eliminates the need for expensive moulds required by the RTM and allows replacement
of the upper mould with a flexible polymeric bag. Hence, VARTM is sometimes referred to as
Resin Infusion under Flexible Tooling or RIFT. Other generic terms given for VARTM include
Vacuum Infusion (VI), Vacuum Infusion Moulding (VIM) and Vacuum Assisted Resin Infusion
Moulding (VARI/VARIM) [6].
Since the development of VARTM processes, researchers and manufacturers have focused on
improving the resin infusion characteristics and processing methods to further reduce costs and
improve part quality. This has led to the development of many materials and improved VARTM
techniques for tooling setup, resin distribution and complete resin impregnation. Some
important developments are outlined in Table 3.2. Many others combined new materials with
innovative process modifications, while attempting to bring out solutions for existing challenges
with conventional VARTM methods. The following sections consider some important VARTM
process variants that originated from research and development work carried out worldwide,
over the last two decades.
CHAPTER 3. REVIEW OF LCM PROCESSES 24
![Page 39: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/39.jpg)
Table 3.2: New materials and technologies developed to improve VARTM process
Research Area Matrix/Core/ Reinforcement Materials
Resin Distribution
Flow Simulation
Tooling/ Vacuum Bagging Supplies
New Developments > Matrix: Epoxy, polyester and vinyl-ester resins available with viscosities below 200
cps and wide range of working times and curing temperatures. Suppliers Include: PTM&W Industries Inc.-Epoxy-N-Fusion®, Gougeon Brothers Inc-West System®, Tri-Tex Co. Inc. - East System®, Polymer Tooling Systems Inc. - Renlnfusion® Epoxies, Endurance Technologies Inc. -Endurance Epoxy Systems, Applied Poleramics Inc.
> Woven Fabric Reinforcement: Carbon, Kevlar, Glass, Aramid or hybrid fabrics available in different densities and variety of fabric styles for improved drapability.
> Reinforcements are also available in roving, mat, knit and 3-D braided forms [6,10]. > Non-Crimp (NC) fabrics with or without tackifiers to facilitate easy layup [6]. > Suppliers of reinforcement materials include The Saertex Group, Metyx Composites,
Hexcel Corporation, Polynova™ Composites., Vectorply Corporation, and JB Martin > Variety of core materials available for infusion (Chapter 5). > Low cost polyethylene based distribution media available for high and low
temperature processing. (Airtech: Resinflow 60, Resinflow 90 HT, Delstar Naltex®) > High Permeable Layers (HPL) for faster and quality infusion. Examples include:
Airtech N4, Airtech N10, Richmond A3000, Colbond EnkaFusion. > Reusable breather and resin distribution medium: Airtech Breatherflow 60. > Inter-laminar infusion for processing thick laminates by VARTM: Using 3-D spacer
fabrics engineered as distribution media or using Lantor Soric® flexible core. > Channel-ln-Bag (CIB) Infusion: Resin conduits integrated into reusable silicone
vacuum bags for resin distribution [14]. > Channel-ln-Core (CIC) Infusion: Resin channels grooved into foam cores to
distribute the resin into the part [15]. > Liquid Injection Moulding Simulation (LIMS): VARTM/RTM flow simulation software,
developed by the University of Delaware- Centre for Composite Materials. > PAM-RTM 2004 LCM simulation software: Flow simulation and process optimization
software for RTM and VARTM type processes developed by the ESI Group. > Simulation based Liquid Injection Control (SLIC): A software package that integrates
flow simulation, setup optimization and process automation and control to provide optimum LCM design and manufacturing solutions; developed at the University of Delaware- Centre for Composite Materials.
> RTM-Worx: Flow simulation software capable of simulating and optimizing RTM/VARTM type processes, developed and licensed by Polyworx Corporation.
> Disposable/reusable vacuum bags available in various levels of conformability (i.e. Econolon®, Stretcholon®, Multibag®).
> Peelplies: Polyester or nylon based, available for low and high temperature processing; silicone or Teflon coated peelplies are also available for easier release.
> Resin infusion lines: Disposable spiral tubing and reusable Omega Flow Lines. > Spray adhesives for holding fabrics together during layup (i.e. Airtac 2, Econotac). > Airtech Fusiontac: A preform tackifier compatible with polyester/vinyl ester resins > Airtech TackStrip: An adhesive coated mesh for holding fabrics during layup- This is
an alternative to spray adhesives developed to prevent laminate contamination. > Suppliers of VARTM tooling materials include Airtech International Inc., Aerovac
Systems Inc., Advanced Composite Materials Inc., and Torr Technologies Inc.
CHAPTER 3. REVIEW OF LCM PROCESSES 25
![Page 40: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/40.jpg)
3.1.4. Seemann Composite Resin Infusion Moulding Process (SCRIMP™)
SCRIMP™ is a VARTM process variant developed in late 1980's by Bill Seemann and patented in
1990 by TPI technologies. SCRIMP™ was developed to reduce infusion times for large and
complex parts and thus improve the overall production rate. The sequence of steps involved in
SCRIMP™ is similar to VARTM except that TPI holds the patent for the particular flow media
and flow process, thus being able to charge for using this particular infusion method [3,16].
3.1.5. Vacuum Assisted Process (VAP®)
The Vacuum Assisted Process (VAP*) was developed by EADS Deutschland and is protected by
several worldwide patents for low cost manufacturing of primary aircraft structures. This
process is characterized by inclusion of an additional Gore Composite Manufacturing (GCM)
membrane, to the conventional VARTM setup, as shown in Figure 3.5. The GCM membrane is a
selectively permeable material that acts as a barrier to resin but is highly permeable to air.
Breather / Flow media Chamber 1 fVaeuuml V a c u u m f o j | Row media Membrane
Vacuum „JU
L! &&
Resin Injection fine
CFC- Preform
Vacuum r
Tooling Sealant Tape
lama si".
*S?SS?^S;£H-J
A
Figure 3.5: Schematic of the EADS VAP8 before (top) and after (bottom) infusion [10]
CHAPTER 3. REVIEW OF LCM PROCESSES 26
![Page 41: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/41.jpg)
Adding this membrane over the distribution medium of a conventional VARTM setup leads to
application of more uniform vacuum pressure on the laminate throughout the infusion process.
This is claimed to result in uniform part thickness, low void content, minimum potential for dry
spot formation and improved overall part quality [10,17].
3.1.6. Fast Remotely Actuated Channelling (FASTRAC)
The FASTRAC process was first developed at the United States Army Research Laboratory (US-
ARL). Instead of using a flow medium to accelerate the resin flow, this innovative process uses a
low cost rigid or flexible tooling in a double vacuum bag setup, in order to create preferential
resin flow channels along the part. A basic FASTRAC infusion setup is shown in Figure 3.6.
Secondary FASTRAC vacuum bag, FASTRAC Non-contacting Tool
Figure 3.6: Schematic of the FASTRAC process [18]
In this process, the preform is placed under a primary vacuum bag and kept under full vacuum
pressure (close to -30 inHg). Then a secondary vacuum bag featuring a non-contacting FASTRAC
tool is placed over this primary vacuum bag and sealed to the open mould surface. Vacuum is
CHAPTER 3. REVIEW OF LCM PROCESSES 27
![Page 42: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/42.jpg)
drawn inside this secondary vacuum bag to keep the FASTRAC tool under full vacuum pressure.
To create the highly permeable resin channels, vacuum pressure inside the primary vacuum bag
is released to relax the bag for a moment and then drawn again to full vacuum pressure. This
allows the primary vacuum bag to form into the FASTRAC tool and create highly permeable
channels that preferentially distribute the resin over the part. Once the part is fully saturated
with resin, the FASTRAC layer can be removed and the part can be cured [18].
3.1.7. Controlled Atmospheric Pressure Resin Infusion (CAPRI)
CAPRI is a VARTM process variant developed and patented by the Boeing Corporation. As
shown schematically in Figure 3.7, CAPRI process includes two adjustments to conventional
VARTM methods, aimed at improving the fibre volume fraction and reducing the thickness
gradient of the final part.
Vacuum Distribution b a 9 media Preform Resin inlet
Figure 3.7: Schematic of the CAPRI process [19]
First, the dry preform is subjected to cyclic debulking under vacuum pressure, after bagging but
prior to infusion. This is said to improve the fibre nesting, reduce the thickness of the laminate
CHAPTER 3. REVIEW OF LCM PROCESSES 28
![Page 43: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/43.jpg)
and hence, improve the overall fibre volume fraction. Additionally, the infusion bucket
containing the resin is kept under partial vacuum using a secondary vacuum pump, while the
outlet lines are kept under full vacuum. The partial vacuum applied to the resin chamber,
though increasing the process complexity and infusion times, reduces the thickness gradient
observed in conventional VARTM processing.
3.1.8. Double Bag VARTM
The Double bag VARTM process is an improved version of conventional VARTM developed and
patented by the Boeing Company, to solve the bag relaxation and thickness gradient formation
associated with conventional VARTM. This process adds a second vacuum bag separated by a
layer of breather cloth to the conventional VARTM setup, as shown in Figure 3.8. The second
vacuum bag applied around the part acts as a caul plate preventing the inner bag from relaxing
behind the flow front. This added bag also restrains the first bag from stretching during and
after the infusion, which helps in maintaining the vacuum integrity. The double bagging process
is expected to reduce the thickness gradient, improve fibre volume fraction and hence the
overall part quality [12].
To vacuum port
Resin feed/
'"'et line Breather cloth
A\..: /_(*>_ i—r —ra^" 1^^™"^^^T. ^^^' —r
Outlet Line: to vacuum port
Figure 3.8: Schematic of double bag VARTM
Outer vacuum bag
CHAPTER 3. REVIEW OF LCM PROCESSES 29
![Page 44: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/44.jpg)
3.1.9. Advanced VARTM (A-VARTM)
The A-VARTM process is a variation of VARTM developed jointly by Mitsubishi Heavy Industries,
Ltd. and Toray Industries, Inc. for fabrication of aircraft primary structures. This process
combines traditional VARTM methods with the addition of preform hot-compaction prior to
infusion and resin bleed-off after infusion, in order to obtain nearly 60% fibre volume fraction.
Enhanced unidirectional properties are attained with the use of advanced Non-Crimp Woven
(NCW) fabrics (Figure 3.9). Primary aircraft structures are manufactured as integral components
by applying co-bonding and advanced pre-forming techniques. Figure 3.10 shows an outline of
the A-VARTM process [20].
Figure 3.9: Outline of NCW fabric [20]
CHAPTER 3. REVIEW OF LCM PROCESSES 30
![Page 45: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/45.jpg)
Carbon fiber | Preparation of tools
( Weaving )
NCW
G I
Hot compaction
( Resin tnfusion.'bleed
( First cunng )
( Post curing J
VaRTM composite
Base resin Hardener X
/ Measured N /Measured I volume J I volume
f De-foam )
I f Mixing j
Figure 3.10: Sequence of steps in A-VARTM process [20]
3.1.10. Single Line Injection (SLI)
SLI is a cost-optimized Liquid Resin Injection (LRI) process to manufacture high quality
composite components in the autoclave. Developed at the German Aerospace Centre (DLR)
Institute for Structural Mechanics, the SLI process combines the advantages of the cost efficient
materials and liquid resin injection processes to manufacture high quality parts in an autoclave.
A schematic illustration of SLI is shown in Figure 3.11 and Figure 3.12.
pressure reducing valve
resin container
vacuum system
resin transfer line
fibre preform
single-sided tool
Figure 3.11: Illustration of SLI process [21]
CHAPTER 3. REVIEW OF LCM PROCESSES 31
![Page 46: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/46.jpg)
P ~ P Pressure P S P 1 autoclave • injection r l g M u l g ' autoclave/ • injection
ffl__ ____— Reducing-Valve
Injection phase
K>
p p ' injection injection
autoclave . ^ k / \ _ _ / ^ _ _
Adjustment of fibre content
Figure 3.12: Pressure distribution during and after resin injection in SLI process [21]
SLI simplifies the traditional VARTM and RTM setups by utilizing a single resin transfer line for
both evacuation of preform and resin injection. This is accomplished by adjusting the pressure
accordingly at different stages as shown in Figure 3.12. Prior to resin injection, the autoclave is
set at pressure level higher than full vacuum pressure while the mould cavity with fibrous
reinforcement is kept under vacuum pressure. Then, resin is injected into the part at a pressure
level equal to that of the autoclave pressure in order to impregnate the part. Following
complete resin impregnation, the pressure inside the autoclave is increased until the desired
fibre volume fraction (typically 60%) is attained and excess resin is drawn out of the resin feed
line with the assistance of the vacuum system. Depending on the part geometry and size, the
placement of the resin feed line can be optimized [21].
3.1.11. High Performance VARTM (Hyper-VARTM™/ Hyper-RTM™)
Hyper-RTM™ and Hyper-VARTM™ are innovative LCM technologies patented by V Systems
Composites Inc. (VSC), San Diego, CA (US). These technologies incorporate the resin
distribution system into the process' proprietary tooling technology, which allows controlled
resin propagation along both in-plane and out-of-plane directions relative to the tool surface.
CHAPTER 3. REVIEW OF LCM PROCESSES 32
![Page 47: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/47.jpg)
Their tooling incorporates the so-called 'port runner devices' that are claimed to be universal,
modular, reusable and provide a non-directional, high permeable system to facilitate the resin
infusion. The result as claimed in the patent (US 6,964,561 B2, Nov 15, 2005), is consistent and
high quality infusion with minimized setup labour and reduced potential for rework or scrap
[22].
3.1.12. Vibration Assisted Liquid Composite Moulding
In this LCM method, either the mould or the resin stream is oscillated using electromagnetic
shakers or motors during the infusion. Such vibrations assisted methods have been shown to
reduce void contents in the final part, enhance flow rates and improve the fibre wetting [23].
3.1.13. Closed Cavity Bag Moulding (CCBM)
CCBM is a relatively new process for low cost manufacturing of FRP composites, patented by
Arctek Inc. It utilizes a silicone based elastomeric material to manufacture flexible vacuum bags
that are form fitted to the shape of the mould [24]. Once the bag is manufactured, the process
follows conventional VARTM methods, as shown in Figure 3.13.
{Photo courtesy of Progress Plastics & Compounds Company Mlsslssougo. ON. Canada, 2008}
Figure 3.13: Typical CCBM procedure
CHAPTER 3. REVIEW OF LCM PROCESSES 33
![Page 48: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/48.jpg)
CCBM bags offer good vacuum integrity and better surface finish compared to traditional
disposable vacuum bags. A properly made CCBM bag can last up to 1000 manufacturing cycles,
leading to potentially significant cost savings in material and labour for small to medium
production runs [24].
CCBM systems are typically supplied in sprayable or brushable forms. Sprayable CCBM systems
generally involve an initial equipment cost, hence are somewhat expensive at the outset
compared to the brushable CCBM systems that only require brushes and squeegees for
manufacturing the bag. Commercially available CCBM type systems include, SWORL™,
Sprayomer Elastomer™, Airtech Multibag™ and Vacuspray™.
3.1.14. Co-Injection Resin Transfer Moulding (CIRTM)
Co-Injection Resin Transfer Moulding (CIRTM) is an LCM method developed jointly by University
of Delaware's Center for Composite Materials (UD-CCM) and Army Research Laboratory (ARL).
In this process, manufacturing of hybrid composites in a single step is achieved by simultaneous
injection of multiple resins into a multi-layer preform. This method can be utilized with
conventional VARTM setup as shown in Figure 3.14.
CHAPTER 3. REVIEW OF LCM PROCESSES 34
![Page 49: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/49.jpg)
phenolic J2027
vinyl ester 411-350
flow direction
vacuum line
SOO<>0<i<X>iJ<M^O<Vg»g o /
distribution media
separation layer
window
mold surface
vacuum bag
release ply
distribution media
Figure 3.14: Schematic of the CIRTM method [25]
The hybrid structure is facilitated using a separation medium, which isolates the individual
resins during infusion but forms a good structural bond upon cure. CIRTM results in substantial
cost savings by the eliminating the need for multi-step processing and secondary bonding when
manufacturing a hybrid composite structure [25].
3.1.15. Euro Composites® Honeycomb Liquid Moulding (EC-HLM)
EC-HLM is a process in which sandwich constructions with honeycomb cores can be produced
using the LCM method, without filling the honeycomb cells with resin. This is achieved by
integrating a unique barrier layer into the fabric layup process as shown in Figure 3.15. The part
is then vacuum bagged as shown in Figure 3.16. In an integrated infusion process, pre-curing of
CHAPTER 3. REVIEW OF LCM PROCESSES 35
![Page 50: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/50.jpg)
the barrier-core-bond takes place in an oven first. This is followed by resin infusion and final
cure. The process is claimed to produce good quality, out-of-autoclave honeycomb structures
[26].
Barrier
Dry fabric/ fiber pack
material: Honeycomb
Figure 3.15: Details of the part during layup [26]
T/"
Resin Vacuum pump
Resin inlet vent
Vacuum Bag
Draining medium
Part in mold after layup Mold
Figure 3.16: Schematic of the EC-HLM process [26]
CHAPTER 3. REVIEW OF LCM PROCESSES 36
![Page 51: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/51.jpg)
3.1.16. VacFlo® Process
The VacFlo® process was developed and patented by Scott Bader Company Ltd., for
manufacturing vacuum infused parts with mould-finished surfaces on both sides. This is
accomplished with the use of matched two-part moulds having a wider flange area that seats a
double-vacuum seal arrangement. During infusion, a first vacuum source is used to draw
vacuum through the flange seal and thus clamp the moulds, while a second vacuum source
facilitates resin flow into the part. The moulds are light weight and typically manufactured using
low cost, fibreglass reinforced laminates. The resulting parts are of same quality as those
produced by VARTM, but show excellent surface finish on both sides [6].
3.1.17. Light RTM
LRTM processes combine certain RTM and VARTM principles to produce vacuum infused parts
with mould-finished surfaces on both sides. This process involves low-cost, two-part moulds
extended with flange-seals, a vacuum source, a low pressure resin injection pump and a
pressure control unit. The vacuum source serves to clamp the moulds and facilitate resin flow
into the part. Resin is introduced into the part using the injection pump, which pushes resin
into the part at pressures that do not exceed the clamping pressure of the mould cavity. In a
typical LRTM setting, the flange seal is clamped with full vacuum pressure (-14.7 psi gauge),
while the mould cavity is maintained at approximately half-vacuum (-7.5 psi gauge). Resin
delivery pressures are constantly controlled during infusion to prevent it from overpowering
the clamping pressure. The push-pull resin infusion method results in complete resin
consolidation of the part in shorter infusion times. This quality infusion comes with a slightly
higher cost than conventional VARTM methods, but the use of low cost moulds and low
CHAPTER 3. REVIEW OF LCM PROCESSES 37
![Page 52: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/52.jpg)
pressure equipment makes LRTM relatively inexpensive compared to traditional RTM methods.
Commercial suppliers of controlled LRTM systems include Magnum Venus Plastech (Lite-RTM),
and JHM Technologies (Zero Injection Pressure (ZIP)-RTM) [6,26,27].
3.1.18. Resin Infusion between Double Flexible Tooling (RIDFT)
RIDFT is an innovative LCM process developed to reduce the infusion times associated with
conventional VARTM. Figure 3.17 illustrates the steps involved in the RIDFT. This process
begins with cutting of the dry preform to fit the desired mould shape. The preform is then
placed in-between two flexible silicone membranes and sealed around the edges. Vacuum is
then pulled between the flexible membranes and the preform is infused while remaining in flat
configuration, as shown in Figure 3.17. The impregnated preform is then vacuum-formed to the
shape of the mould by sealing and evacuating the cavity between the bottom silicone layer and
the mould. Finally the part is cured and removed from the mould. Because the preform is flat
during resin infusion, problems associated with wetting out complex geometry components are
eliminated. Since the part stays over the flexible silicone layer at all times, less cleanup and pre-
manufacturing mould preparation is required with each manufacturing cycle. This process is
claimed to produce parts with low void content, at low emissions of Volatile Organic
Compounds (VOCs) and lower tooling costs compared to conventional RTM methods. However,
parts that can be manufactured with this method are of somewhat limited geometric
complexity and size, due to the vacuum forming aspect of this process and the largest available
machine size (10 ft x 15 ft x 4 ft height- Figure 3.18) [29].
CHAPTER 3. REVIEW OF LCM PROCESSES 38
![Page 53: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/53.jpg)
,j. ^
Step 1: Load fiber Step 2: Seal Resin Infusion Bag
Step 3: Resin Infusion
L
Step 4: Seal Vacuum Chamber
1 Step S: Vacuum
Form Part Step 6: Demold
Figure 3.17: Schematic of the RIDFT process [10]
Figure 3.18: Industrial RIDFT machine 10 ft x 15 ft x 4 ft [29]
CHAPTER 3. REVIEW OF LCM PROCESSES 39
![Page 54: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/54.jpg)
3.1.19. Flexible Injection
Flexible Injection is a new, patent pending LCM technology, developed and currently being
researched at Ecole Polytechnique de Montreal, in collaboration with General Motors (GM) of
Canada Ltd. Flexible Injection can produce parts much faster than other LCM processes without
compromising the part quality. The major steps involved in the Flexible Injection process are
illustrated in Figure 3.19 [30].
The process begins with the placement of fibrous reinforcement into the injection chamber.
Then a flexible membrane is placed over the reinforcement stack and sealed around the mould
(Figure 3.19 - Step 1). Full or partial vacuum may be applied at one extremity of the injection
chamber causing the flexible membrane to compress the reinforcement stack as shown in step
2. The required amount of resin is injected under pressure into the mould cavity, which fills the
fraction of the mould cavity closer to the injection port as shown in step 3. The resin injection is
immediately followed by injection of a non-reactive compaction fluid under pressure, into the
upper mould chamber. This step accelerates the resin flow and impregnation into the fibrous
reinforcement. After complete impregnation of resin into the reinforcement stack, resin vents
are closed and the part is allowed to cure at the desired temperature and pressure. Once the
part is cured, the compaction fluid is drained out of the cavity and the part is demoulded. The
initial process development work demonstrates that the Flexible Injection technology offers
faster and quality resin infusion [30].
CHAPTER 3. REVIEW OF LCM PROCESSES 40
![Page 55: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/55.jpg)
Compaction Quid ' outlet
Compaction fluid inttet
Step 1. Mould Setup
A - * *
Step 2. Vacuum Application
< ompartio* Hiiirt Compaction Fluid
Closed
r ^ ^ j Impregnated
I I » . T
Step 3. Resin Injection s t eP 4- Homogeneous compaction of impregnated fibres
Figure 3.19: Flexible Injection Process [30]
3.1.20. Resin Film Infusion (RFI)
RFI uses solid resin film or so called 'prepreg resin' to saturate the dry preform with resin. In the
basic RFI process, the tool (mould) is covered with resin film of required thickness and dry
preform is laid-up on top of this resin film as shown in Figure 3.20. The mould is then bagged
with disposable materials. Subsequently, vacuum or autoclave pressure is applied to the setup,
in conjunction with an appropriate thermal cure cycle. Upon curing, the resin viscosity lowers,
allowing the resin to diffuse into the structural preform. With the RFI setup, the resin only
infuses through thickness, providing considerably shorter infusion times versus traditional LCM
CHAPTER 3. REVIEW OF LCM PROCESSES 41
![Page 56: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/56.jpg)
processes. However, RFI requires high temperature rated moulds and at least an oven for
curing, which makes this process somewhat expensive compared to conventional LCM methods
[31].
Bagging film N10 breather
Al tool \
Sealing tape
Bleeder
Perforated film Flashbreaker
— Vacuum
Resin film
Dry preform Cork dam
Airpad caul plate with bleed-out holes
Figure 3.20: Schematic of the RFI process [31]
3.1.21. Semi-Preg Infusion
In this form of resin infusion, reinforcements partially impregnated with resin are used along
with standard vacuum bagging methods and oven curing to produce the final part. These so
called 'sermi-pregs' or out-of-autoclave prepregs are low cost alternative to standard prepregs.
They exhibit shelf life of up to one year at room temperature. Commercially available semi-preg
systems include [6]:
> Advanced Composites Group ZPREG o Resin stripes on one side of fabric
> Cytec Carboform o Resin impregnated random mat between the two fabric layers
> SP Systems SPRINT® (SP-Resin Infusion New Technology) o Resin between two fabric layers
CHAPTER 3. REVIEW OF LCM PROCESSES 42
![Page 57: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/57.jpg)
3.2. Suitability LCM Processes for Mouldless Manufacturing The problems of vacuum leaks and foam core distortion under vacuum pressure experienced
during previously implemented mouldless VARTM method (Chapter 2) can be addressed in two
different ways. The first way is to use more robust bagging material and precise alignment
fixtures for manufacturing. The second way is to select and apply a process variant that readily
solves these problems.
The use of robust bagging material and precise alignment mechanisms will slightly increase the
cost of mouldless manufacturing. Though this is undesirable, the added cost will likely result in
better part quality. On the other hand, using an available technology usually comes with a
licensing fee. For some processes developed with the intention of reducing costs, this licensing
fee is usually small and is affordable by small or medium size companies. Thus, a process
variant is worth considering if it promises to solve the problems encountered in mouldless
VARTM.
VacFlo, Light RTM, RFI, and Semi-preg Infusion methods come with relatively high equipment
and material costs compared to conventional VARTM. Hence these processes are not
considered any further for mouldless manufacturing. Other processes such as VAP and Double
Bag VARTM promise better vacuum integrity as compared to conventional VARTM, but will
suffer similar core distortion under full vacuum pressure. Distortion of foam core under vacuum
pressure is a direct consequence of excessive 'pleats' (Figure 3.21) in the vacuum bag, which
conform onto the part unevenly when vacuum is applied to the part. Such pleats are
unavoidable when bagging complex parts with tough, disposable vacuum bags. These methods
will also have to rely on precise fixtures for controlling the part dimensions, which is somewhat
CHAPTER 3. REVIEW OF LCM PROCESSES 43
![Page 58: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/58.jpg)
difficult to achieve when the part is surrounded with uneven vacuum bag pleats. This leads to
the conclusion that a vacuum bag that is form fitted to the shape of the part, will provide a
better solution to part shape retention in mouldless manufacturing. A form fitted vacuum bag
will allow application of uniform pressure around the entire tool, which may help minimize the
deflection of the foam core under vacuum pressure experienced in previously implemented
mouldless VARTM. CCBM is therefore a potential alternative to VARTM for mouldless
manufacturing. The competitive advantages of the CCBM process are listed in Table 3.3.
Table 3.3: Advantages and Disadvantages of CCBM process [24]
Advantages Reusable
Repairable
Many options for sealing the bags Translucent
Low cost mould requirements Less wastage
Material compatibility
Integrated manufacturing Workable
Robust
Suitable for all part sizes and geometry Better part quality
Description Tough and self cleaning bags that can last over 1000 cycles without any mould release application; improves the process repeatability. CCBM bags are easily repaired and restored to original condition when damaged by accident or rough handling. Unlike disposable vacuum bags, CCBM bags can be sealed in many different ways (i.e. flange seal, zipper seal etc.). Allows the operator to visually monitor the flow front and control the infusion. Existing moulds can be built up at low cost to make the mould for CCBM bag fabrication. CCBM reduces wastage of resin and bagging consumables compared to conventional VARTM. CCBM is compatible with almost all available matrix and reinforcement systems (i.e.: carbon-epoxy, fibreglass-polyester). CCBM allows manufacturing with integrated cores and ribs without the need for complex vacuum bag pleating. Minimal odour, no VOC's, easy storage and handling, and readily conforms to the shape of the mould. CCBM infusion offers improved vacuum integrity, reduced risk of bag leaks and eliminates the infusion complexities such as resin rich or resin starved areas caused by disposable bags. As the complexity and size of the part increases, CCBM makes vacuum bagging easier and cost efficient for low production quantities. CCBM can create cosmetically attractive bag side surface with no wrinkles.
Disadvantages
• Relatively high initial costs compared to traditional VARTM with disposable materials. • The durability of the bags is heavily dependent on the quality of bag manufacturing.
CHAPTER 3. REVIEW OF LCM PROCESSES 44
![Page 59: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/59.jpg)
Figure 3.21: A part bagged with disposable vacuum bag [32]
CCBM bagging for the fuselage requires a multi-section bag, with a sealing mechanism
adjustable to tightly seal the bag around the entire part. An additional requirement is such that
the selected CCBM system and sealing method should be economical for part counts below 10,
in order to realize the economic advantages of mouldless manufacturing. From the review of
current CCBM techniques, it becomes apparent that this process is intended to be used for
manufacturing with rigid moulds, and hence can be expensive at the outset. Following the
detailed review of LCM methods, the CCBM process was selected for demonstration of
mouldless manufacturing; however, it was clear that significant process development and
assessment would be required to determine the most efficient and economical mouldless
CCBM setup.
CHAPTER 3. REVIEW OF LCM PROCESSES 45
![Page 60: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/60.jpg)
CHAPTER 4. CCBM PROCESS DEVELOPMENT
This chapter presents process development work carried out to develop a suitable CCBM
technique for mouldless manufacturing. Various methods of bag-sealing and resin distribution
for CCBM process were identified. A series of manufacturing trials were conducted using flat
CCBM bags to determine the suitable techniques for mouldless manufacturing. Finally a Process
Value Analysis (PVA) was performed to evaluate the feasibility of CCBM for mouldless
manufacturing.
4.1. Mouldless CCBM manufacturing considerations
Applying CCBM in mouldless manufacturing setting requires a suitable bag sealing method that
does not depend upon rigid flanges. Additionally, efficient resin distribution techniques and
economic bag manufacturing strategies, if developed, would lead to further cost savings; hence
would make CCBM economical for mouldless manufacturing in low production quantities.
Several bag-sealing and resin distribution concepts for mouldless CCBM manufacturing,
investigated as part of this research are described in the following sections.
4.1.1. Sealing Mechanisms for Mouldless CCBM
Conventional CCBM process often uses extruded silicone sections, such as the Flexseal "D" to
create the vacuum seal. Some of the sealing configurations commonly used in the CCBM
process are shown in Figure 4.1. With the exception of interlocking seals, all seals require a rigid
mould surface and a flange as shown in Figure 3.13. This leaves the interlocking seal as the
most suitable sealing configuration for mouldless CCBM. However, this sealing method requires
the two parts of the seal to be positioned accurately on the top and bottom halves of the bag to
form an air-tight enclosure. This is somewhat difficult to attain considering the size and
CHAPTER 4. CCBM PROCESS DEVELOPMENT 46
![Page 61: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/61.jpg)
geometric complexity of the fuselage. Hence an alternative, more economical sealing method
recommended in [34] was investigated. In this method, a tape with silicone based adhesive
backing (i.e. Airtech Flashbreaker® or Teflease®) is permanently bonded to the CCBM bag. The
taped surface of the bag is sealed against the tool using disposable sealant tape. This sealing
technique eliminates the high initial costs associated with the extruded silicone seals.
seal seal seal 16mm
l i I
Photo courtesy of Airtech Advanced Materials Group, Rubber Silicone Seals, [Online Catalogue], 2009, [Cited 12 Oct 2010], Available: http://catalogue.airtech.lu/product.php7product id=29&lane=EN
Figure 4.1: Common reusable seal configuration for CCBM
4.1.2. Resin Inlets and Outlets
One way to reduce the material waste associated with traditional VARTM is to use reusable
resin infusion lines. In the CCBM process, this is accomplished by permanently moulding the
resin infusion channels into the vacuum bag. To do this, two options were considered. The first
option was to mould the shape of tube into the bag by placing a waxed polymer tube on the
tool and fabricating the bag over it. The region of resin channels can be locally reinforced with
multiple layers of silicone to provide the necessary vacuum integrity. If this method was to
prove unsuccessful, a more conservative second option can be considered, in which extruded
CHAPTER 4. CCBM PROCESS DEVELOPMENT 47
![Page 62: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/62.jpg)
silicone Omega Flow Lines (OF313) supplied by Airtech Corporation would be permanently
embedded into the CCBM bag during manufacture.
4.1.3. Resin Distribution in CCBM
The resin can be distributed across the part using either a disposable resin distribution medium
or by means of resin channels that are moulded into the CCBM bag (CIB infusion). The use of a
distribution medium is more conventional and has been proven to work in conventional VARTM
or CCBM settings [3,24]. On the other hand, CIB infusion is specifically advantageous for large,
complicated parts as it eliminates material and labour costs associated with the use of
disposable distribution medium. Additionally, the placement of such resin distribution channels
can be optimized to shorten resin infusion time and reduce resin consumption.
4.2. CCBM Manufacturing Trials
Three flat CCBM bags were manufactured to evaluate the functionality of various bag sealing
and resin distribution techniques discussed in section 4.1. The procedure employed in these
manufacturing trials and their outcomes are discussed in the following sections. Supplier
information of the materials used for this manufacturing trial is provided in Appendix A.
Arctek CCBM system [35] was chosen for these manufacturing trials, as it was the least
expensive, brushable CCBM system available. The required materials can be purchased in small
quantities. Important components of Arctek CCBM system are shown in Figure 4.2. It consists of
Proflex NS® silicone, which is supplied in 850 ml cartridges. This is a single-part, atmospheric-
moisture cure silicone. For 0.025 in. thick layer of Proflex NS®, the cure time is approximately 1
hour at 77° F/ 50% RH. An Arctek CCBM bag is fabricated by applying several layers of Proflex
CHAPTER 4. CCBM PROCESS DEVELOPMENT 48
![Page 63: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/63.jpg)
NS® silicone over a rigid mould. A layer of Confortex® fabric is applied in-between the silicone
layers, which increases strength and tear resistance of the CCBM bag. Technical details and
manufacturing methods of this product can be found in [24].
Training DVD
Flexseal'D' reusable seal with adhesive backing
Proflex NS® silicone cartridges
Confortex® fabric
Figure 4.2: Components of Arctek CCBM system [35]
4.2.1. CCBM Manufacturing Trial # 1
In this trial, a 9 in. x 20 in. CCBM bag was manufactured for infusing a flat panel with the
assistance of disposable distribution medium. The bag featured Airtech Flashbreaker tape
bonded around the perimeter to facilitate sealing using disposable sealant tape. Half-circular
profiles were moulded into the bag to create resin inlet and outlet as shown in Figure 4.3. The
manufactured CCBM bag is shown in Figure 4.4. Step by step manufacturing method is provided
in Appendix B.
CHAPTER 4. CCBM PROCESS DEVELOPMENT 49
![Page 64: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/64.jpg)
Flashbreaker tape
Application of first silicone layer
Half-circular profile to create resin inlet/outlet lines
Figure 4.3: CCBM bag manufacturing trial #1: tool setup
Top Surface
p
Resin Flashbreaker tape Inlet/Outlet
Flexible aluminum frame
Silicone tubes bonded to
reinforce the resin channels
{
Figure 4.4: CCBM bag manufacturing: trial #1 results
The results showed that using Airtac 2 spray adhesive to hold the Flashbreaker tape against the
tool (Appendix B) does not work for this sealing mechanism. Indeed, the spray adhesive
contaminated the surface of the tape, leading to permanent adhesion of the sealant tape to the
bag. Further, the half circular resin channels moulded into the bag collapsed under the applied
vacuum pressure. This bag was recovered by attaching a flexible aluminum frame around the
perimeter and bonding silicone tubes to locally reinforce the resin channels, as shown in Figure
CHAPTER 4. CCBM PROCESS DEVELOPMENT 50
![Page 65: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/65.jpg)
4.4. This CCBM bag used Airtech Resinflow 60, disposable resin distribution media to speed-up
the infusion process.
4.2.2. CCBM Manufacturing Trial # 2
A second CCBM bag was manufactured with slightly modified procedure to overcome the
difficulties in trial #1. In this trial, Airtech OF313 omega flow lines were embedded into the bag
to create the resin flow lines, and Teflease tape was bonded along the perimeter of the bag to
form the seal as illustrated in Figure 4.5. During manufacturing of the bag, Teflease tape was
held down to the tool surface with the use of a double-sided tape to prevent surface
contamination. The manufactured CCBM bag is shown in Figure 4.6. The problems encountered
in trial #1 were solved and much better quality bag (Figure 4.6) was produced in trial #2. Resin
infusion was facilitated with disposable resin distribution media. This bag is still in good
condition after 10 manufacturing cycles.
OF 313 Omega Flow line S> Resin Inlet/Outlet
Figure 4.5: CCBM bag manufacturing trial #2: tool setup
Teflease tape
CHAPTER 4. CCBM PROCESS DEVELOPMENT 51
![Page 66: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/66.jpg)
I Top Surface
\ Teflease tape Resin Inlet/Outlet
OF 313 Omega Flow Line
Figure 4.6: CCBM bag manufacturing: trial #2 results
4.2.3. CCBM Manufacturing Trial # 3
A third manufacturing trial was carried out to determine the feasibility of moulding resin
distribution channels into the bag. In this attempt, 0.125 in. diameter wire wax was placed onto
the flat tool using double-sided tape. Mould release wax was then applied on the tool surface
and CCBM bag section was manufactured over this setup, as shown in Figure 4.7.
.125 in. diameter „.„.;.:.•.;.««>:«:«:*:•«••.•::•:«::•:•:•
me wa* c
Tool preparation and
ig
Figure 4.7: Sample CCBM bag section with resin distribution channels
CHAPTER 4. CCBM PROCESS DEVELOPMENT 52
![Page 67: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/67.jpg)
Following this trial, a complete bag section was manufactured to compare the infusion
characteristics and the part quality. Infusions with CIB method and disposable distribution
media are compared in Figure 4.8. In CIB infusion, the resin travels preferentially through the
distribution channels and then into the part, resulting in a flow front pattern as shown in Figure
4.8-(a). Resin infusion with disposable distribution medium results in a linear flow front as
shown in Figure 4.8-(b).
(a) (b)
Figure 4.8: Flow profiles of CIB method (a) and disposable distribution media (b)
4.3. Value Analysis of the CCBM process
CCBM manufacturing trials showed that with appropriate use of materials and low-cost bag
manufacturing techniques, CCBM can be made suitable for mouldless manufacturing of
complex geometry components. In order to determine the most economic CCBM setup for
mouldless manufacturing and assess its feasibility against mouldless VARTM method, a Process
Value Analysis (PVA) was conducted.
CHAPTER 4. CCBM PROCESS DEVELOPMENT 53
![Page 68: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/68.jpg)
4.3.1. Introduction to Value Analysis
A product is considered valuable if it shows excellent performance characteristics and good
physical appearance, relative to its cost. In mathematical terms, the value of a product can be
expressed as [36].
_ (Performance + Capability) Cost { 4 1 }
_ Function Cost
The above expression shows that the value of a product can be increased either by minimizing
the cost or by maximizing the performance. In other words, the most valuable product will have
the highest functional worth (lowest cost to perform a given function). The concept of Value
Analysis was first developed in 1945, by Lawrence D. Miles, an engineer from General Electric
(GE) Company [37]. Mr. Miles found that with a systematic approach and clear understanding
of functional worth of the product, one could meet or improve product performance and
reduce its manufacturing costs. His approach to continuous improvement was called the Value
Analysis. The Value Analysis can be carried out on designs or processes, as an improvement
effort at any stage of a product life cycle. When several process variations exist, the Process
Value Analysis (PVA) is employed to determine the net value of each process variation. The
value of each process variation is then assessed to draw conclusions on the process feasibility.
Potential benefits of PVA include reduced material use and costs, reduced waste, reduced
distribution costs, improved profit margin, and improved customer satisfaction [36,37].
CHAPTER 4. CCBM PROCESS DEVELOPMENT 54
![Page 69: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/69.jpg)
4.3.2. CCBM Process Value Analysis
The first step of the PVA was to identify the processes suitable for mouldless manufacturing
and develop a PVA matrix. Various sealing and resin infusion techniques investigated as part of
the manufacturing trials were assessed, to identify process variants applicable for mouldless
manufacturing (Table 4.1). In this PVA, it is assumed that each method described in Table 4.1
would yield similar part quality. Considering performance, the CCBM bag I, with extruded
silicone seal is easier to install during manufacturing than CCBM bag with sealant tape. This
benefit of the CCBM bag-l is realized at a slightly higher initial cost. In order to express the
functional worth as monetary values, all functional characteristics of the process variants were
grouped in terms of material and labour costs, with an initial labour rate of 20 $/hour.
Table 4.1: PVA matrix: processes for mouldless CCBM/VARTM
Process Variation CCBM bag 1 CCBM bag II CCBM bag III
VARTM
Description CCBM with extruded silicone seal and distribution medium CCBM with sealant tape and distribution medium CCBM with sealant tape and resin distribution channels embedded in the bag Traditional VARTM with disposable materials
Since CCBM bags are reusable for up to 1000 manufacturing cycles, the functional worth of
these processes were determined by estimating the cost per part for increasing part counts.
The total manufacturing cost per part was calculated by adding the fixed and variable costs
associated with the processes. All value assessment calculations were carried out using a
custom Microsoft Excel template created for this PVA. The assumptions made combined with
the equations used and the estimated values of each process are provided in Appendix C.
CHAPTER 4. CCBM PROCESS DEVELOPMENT 55
![Page 70: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/70.jpg)
4.3.3. PVA Results
The total costs of manufacturing the fuselage for increasing part count were plotted at a labour
rate of $20/hour (Figure 4.9). The results show that the process development work, which
evolves to CCBM III, makes CCBM economical for 4 fuselages. This is an improvement from the
published manufacturer's data, which justifies the cost of CCBM for over 5 complex parts in a
conventional CCBM setting [35]. Thus, CCBM appears to be economical for mouldless
manufacturing. In order to assess the sensitivity of these results to labour cost, the cost
estimates were regenerated with a labour rate of $40/hr. The results shown in Figure 4.10,
suggests that doubling the labour rate increases the cost of the first part, but makes CCBM
economical for fewer parts (3 parts). This is primarily due to the significant labour cost savings
offered by CCBM for increasing part counts.
$3,000 -i
$2,500 - -
$2,000 •
o $1,500 — 4-* 0)
z
$1,000 -
$500
$0
0 1 2 3 4 5 6 7 Number of Parts
Figure 4.9: Cost of the fuselage for increasing part count at labour rate of 20$/hr
CHAPTER 4. CCBM PROCESS DEVELOPMENT 56
![Page 71: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/71.jpg)
<po,uuu -
$2,500 -
$2,000 -
** ° $1,500-
z $1,000 -
$500-
$0-
-•-VARTM
— CCBM-I
— CCBM II
-*-CCBM III
3|r ^ " ^
— ~~ ~ z^^^^^^^^"^ — ~
— -
1 1 i 1 i 1
3 4
Number of Parts
Figure 4.10: Cost of the fuselage for increasing part count at labour rate of 40$/hr
4.3.4. PVA Conclusions
Mouldless CCBM is a viable option for manufacturing large complex geometry components.
CCBM III is cost effective after manufacturing 3 to 4 parts. Process robustness, repeatability and
part quality are likely to improve with proper implementation of mouldless CCBM.
CHAPTER 4. CCBM PROCESS DEVELOPMENT 57
![Page 72: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/72.jpg)
CHAPTER 5. FUSELAGE MATERIAL SELECTION
This chapter provides a brief overview of the sandwich theory and discusses the choice of
materials for the new GeoSurv II fuselage.
5.1. Sandwich Theory
Sandwich constructions are widely used in many structural applications of advanced composite
materials. They consist of thin face sheets or skins adhesively bonded to both surfaces of a
relatively thick, low density core material. The core serves to increase the overall laminate
thickness, thereby keeping the Fibre Reinforced Polymer (FRP) skins apart. This leads to a
dramatic increase in flexural rigidity of the laminate for a small added weight. To illustrate the
sandwich principle, consider a symmetric sandwich beam in bending (Figure 5.1).
mffh
4 ¥
-» X
m/m
Figure 5.1: Sandwich beam subjected to three point bend
The stiffness coefficient (D) of this beam is given by [1]:
D, beam = {EI\ beam
= {El)core + {El\ skins
s 6 2 c 12
(5.1)
CHAPTER 5. FUSELAGE MATERIAL SELECTION 58
![Page 73: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/73.jpg)
Where, E is the Young's Modulus of the material and / is the area moment of inertia of the
(d \ beam. For sandwich constructions featuring relatively thin skins — > 5.77 and relatively weak
\t )
core ' * • < > ! «
KE< C'
, the expression for D can be approximated as:
beam i
btd:
(5.2)
The above expression shows that the distance between the skins, "d" has greater influence on
the flexural rigidity of the beam compared to other variables in equation 5.2. The core
effectively increases this distance for a small increase in weight, thus making sandwich
construction stronger and stiffer than the corresponding monolithic counterparts. In bending of
a sandwich beam, the skin bears most of the bending stresses while the core predominantly
carries shear loads (Figure 5.2).
Top skin: In compression
VCore: In shear
^ -»-siV Neutral Axis
Bottom skin: In tension
Figure 5.2: Sandwich beam in bending
The core also serves to distribute the local loads to be carried by the skins over the entire
structure, without causing local failure; this makes sandwich construction an excellent design
solution for components exposed to impact and dynamic loading. The compression strength of
CHAPTER 5. FUSELAGE MATERIAL SELECTION 59
![Page 74: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/74.jpg)
the core prevents local buckling of the skins and the shear strength coupled with good adhesion
between the skin and core aids in holding the skins in place under dynamic bending loads. The
latest core materials also offer excellent heat insulation, acoustic insulation, vibration damping
and fire resistance characteristics [2]. Detailed theory of sandwich structures can be found in
the literature e.g. [39-41].
5.2. Core Material for GeoSurv II Fuselage
The current fuselage uses Celfort® 300 EPS foam as the core material. This choice was primarily
driven by the lower cost of EPS foam and the proof of concept nature of the initial
manufacturing trials (Maley, [3]). Celfort® 300 has a nominal density of 1.42 lbs/ft3 and
compression strength of 30 psi. Initial infusions on test articles revealed that Celfort® 300
degraded due to surface resin absorption. Indeed, the infused sections distorted under full
vacuum pressure about 10 minutes after resin impregnation. Hence, during manufacturing of
the current fuselage, the foam parts were assembled and primed with a layer of fast cure epoxy
(West System 105) prior to infusion. The added resin layer allowed better handling of the core
during manufacturing and slightly increased its flexural stiffness and dimensional stability, at
the cost of increased weight. The manufactured component had up to 0.31 in. (7.8 mm)
deviation from the desired dimension, at several locations [3].
Replacing Celfort® 300 with an aerospace grade, structural core material promises potential
weight savings by eliminating the need to seal the foam surface prior to infusion. Core materials
specially formulated for structural applications and VARTM processing offer good mechanical
properties, lower surface resin absorption and could potentially improve the dimensional
CHAPTER 5. FUSELAGE MATERIAL SELECTION 60
![Page 75: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/75.jpg)
tolerances of mouldless VARTM manufacturing. Hence, an investigation into state-of-the-art
structural core materials was performed to select a suitable core material for the new fuselage.
A wide range of foam, balsa wood and honeycomb core materials are available, with various
densities and finishes; all designed to meet specific structural and manufacturing requirements
of aerospace, marine, wind energy and transportation industries. The choice of core material
for the GeoSurv II fuselage must meet the requirements imposed by the nature of mouldless
VARTM manufacturing, listed in Table 5.1, in addition to complying with the UAS airframe
requirements outlined in Table 2.1. The following sections discuss various candidate structural
foam and balsa core materials that could be used for mouldless VARTM manufacturing.
Table 5.1: Manufacturing requirements for the core material
Requirements for Core Material
1. The core material shall be resistant to vacuum pressure (Compression Strength > 14.5 psi).
2. The core material shall feature nearly 100% closed-cell structure.
3. The core material shall be compatible with epoxy resins.
4. The surface finish of the core material shall lead to excellent skin-core adhesion.
5. The core material shall have excellent specific flexural stiffness to allow for near net shaped manufacturing.
6. The core material shall remain dimensionally stable after machining and during processing.
5.2.1. Structural Foam Core Materials
Structural foam cores are often preferred for Liquid Composite Moulding (LCM) processes. This
is mainly because most foam materials are easily machined or thermoformed. Their surface
finishes and cell packing densities can be tailored to meet a wide range of design requirements.
CHAPTER 5. FUSELAGE MATERIAL SELECTION 61
![Page 76: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/76.jpg)
Also, most foam cores are readily available with optimized grooves and scrim-backings to
enable faster and better resin infusion.
Structural foams are processed from a variety of thermoset and thermoplastic polymers
including Polyvinyl Chloride (PVC), Polyurethane (PU), Styrene Acrylo-Nitrile (SAN), and
Polymethacrylimide (PMI), Polyethylene Terephthalate (PET), Polyester and Polyisocyanurate.
The mechanical properties, density and service temperature of structural foam cores can be
modified significantly by changing the ratios of chemical additives and various process
parameters such as pressure and temperature. The latest manufacturing technologies have
produced foam cores in densities ranging from 2 lbs/ft3 to 50 lbs/ft3, suitable for service
temperatures in the range of -184C to 260 C (-300° F to 500 F). They are available in sheets of
thicknesses up to 4 in. Some structural foam cores are also available in large block form to allow
for components to be machined as single integral bodies [42].
The selection process should carefully consider the mechanical properties, chemical resistance,
toxicity and costs of the available structural core materials. Table 5.2 describes some closed-
cell, semi-rigid foam cores that are potential candidates for the new fuselage. All of these
materials have better overall structural performance than Celfort 300 EPS foam utilized in the
current fuselage. They are available in various grades to meet the UAS certification standards
for fire, smoke and toxicity. The core selection for the new fuselage took into consideration, the
mechanical properties and costs of various structural foam materials, as described in section
5.2.4.
CHAPTER 5. FUSELAGE MATERIAL SELECTION 62
![Page 77: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/77.jpg)
Table 5.2: Structural foam cores suitable for mouldless VARTM manufacturing'
Material Trade Names
Chemical Composition
Manufacturer
Unique Features
Closed-Cell Content
Density Range (lbs/ft3) Continuous thermal stability (° F) Grades
Cross-Linked PVC AIREX-C DIVINYCELL-H KLEGECELL-R Polymer based on cross linked Poly-Vinyl Chloride system • Alcan Composites • Diab Inc.
• Good peel strength • Compatible with most
resin systems • High thermal stability • Thermoformable above
212° F • Low water/resin
absorption • Contain gas under
pressure; possibility of out-gassing over time
> 95%
2 to 15.61
-328 to 158, 284 (High Temp.)
-» AIREX • C70: Universally
structural • C71: Elevated temp. • C52: Industrial
processing -» DIVINYCELL • H: High performance • HT: Aerospace grade • HP: Prepreg processing • F: low FST (Fire, Smoke
and Toxicity) • HCP : High density
Linear PVC AIREX-R
Polymer based on linear Poly-Vinyl Chloride system • Alcan Composites
• More elastic than cross linked PVC
• Good fatigue and impact resistance
• Thermoformable • Slightly lower
mechanical and thermal properties compared to cross linked PVC
>95%
3.75 to 8.70
131
• AIREX R63: Damage tolerant foam
PU LAST-A-FOAM NIDA FOAM PU
Polyurethane based chemical system
• General Plastics • Nida-Core Corp. • Resistant to most
chemicals and solvents
• More brittle and less fatigue resistant than PVCand SAN
• The surface at the resin-core interface tends to deteriorate over time
>95%
2 to 40
275, 320 (High Temp.)
• FR 6700: Aircraft grade
• FR7100: Modelling grade
• FR 10100: High temp.
• FR4300: Formable
• TR: Marine grade
SAN CORECELL
Styrene Acrylo-Nitrile based chemical system • SP Systems
(North America)
• Good machinability
• Resistance to water and fuel oil
• Minimal density variation
• No out-gassing problems
• Compatible with most resin types
• Thermoformable
>95%
3.6 to 19.7
185, 230 High Temp.)
• Core-Cell A: For dynamically loaded structures
• Core-Cell P: Prepreg processing
• Core-Cell T: For decks and interiors
• Core-CellS: sub-sea applications
2 Information provided in this table were obtained from the manufacturers' data. The references are provided in Appendix D.
3 Problem areas are identified in bold and italicized font.
CHAPTER 5. FUSELAGE MATERIAL SELECTION 63
![Page 78: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/78.jpg)
Table 5.2 Continued...
Material Trade Names
Chemical Composition
Manufacturer
Unique Features
Closed-Cell Content
Density Range (lbs/ft3) Continuous thermal stability (° F) Grades
PMI ROHACELL
Polymethacrylimide based chemical system
• Evonik Industries /Degussa
• Resistant to organic solvents such as benzene, xylene and monostyrene
• Resistant to fuel constituents and solvents for paints
• Features better mechanical properties compared to other structural foam cores
• Optimized for LCM processes
• Relatively lower resin absorption than other foams
2.0 to 6.9
320
->Rohacell • RIST: Low surface resin
absorption • RIMA: Finest cells for
minimum surface resin absorption
• IG: Industrial grade • A: Aircraft grade • WF: Heat resistance
grade • XT: Extended temp. • S: Easy to shape and
machine
PET AIREX-T NIDA FOAM PET Poly Ethylene Terephthalate based chemical system • Alcan Composites • Nida-Core Corp.
• Good high temp, stability
• Easily machined and thermoformed
• Chemically stable
> 95%
6.3 to 20
212, 302-392 (High Temp.)
• AIREX-T 90: Easy processing
• AIREX-T 91: Easy processing
• Nida Foam PET100/150: Structural
Polyester AIRCELL
Polymerized cross-linked aromatic polyester system
• Polyumac Inc.
• Good impact and fatigue resistance
• Durable and resilient
• flame retardant, non-friable
Closed content cell comparable to PVC
4 to 36
-320 to 165
• Aircell T: structural
Polyisocyanurate ELFOAM
Polyisocyanurate based chemical system • Elliott
Company
• Excellent chemical resistance and resin compatibility
• Easily machined, perforated and cut
• Class 1 flammability rating
• Good thermal insulation
Closed-cell content comparable to PU 2 to 6
-297 to 298
• Elfoam P: Structural series
CHAPTER 5. FUSELAGE MATERIAL SELECTION 64
![Page 79: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/79.jpg)
5.2.2. Balsa Wood Cores
Balsa wood is another core material known for its high specific compression strength and
stiffness. It is dried and prepared from naturally harvested balsa lumber in the end-grain
configuration, as shown in Figure 5.3. In the end-grain configuration, balsa core features lean
closed cells tightly packed and oriented perpendicular to the plane, forming a closed-cell,
honeycomb like structure at microscopic level. This leads to its ability to resist high compression
and dynamic loads. Apart from this, balsa cores are fire resistant and act as thermal and
acoustic insulators in sandwich constructions [42,44].
Figure 5.3: Balsa wood core-end grain configuration [43]
The main disadvantage of balsa core is its high density, with lowest minimum values ranging
between 5.5 lbs/ft3-6 lbs/ft3. The density factor is further aggravated with balsa's high surface
resin absorption characteristics, making it non-preferable for weight critical applications. The
three main manufacturers of advanced structural core materials, Diab Corporation, Alcan
Composites and Nida Core Corporation are competing to produce lightweight-consistent
CHAPTER 5. FUSELAGE MATERIAL SELECTION 65
![Page 80: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/80.jpg)
density balsa cores by utilizing advanced processing techniques. Balsa cores are now pre-sealed
with primers specifically formulated for lower surface resin absorption, and are available in
sheets of thicknesses up to 4 in. or as contoured blocks held together with fibreglass scrim [44].
5.2.3. Other Core Materials
Other core materials that may be utilized in VARTM applications include the following:
> Cedar wood: It is a natural lumber often used as core material in strip-plank
construction. It features grains running parallel to the plane, offering some stand
alone bending rigidity compared to balsa wood. However, it exhibits poor impact
resistance, low torsional rigidity and compression strength compared to balsa wood
[44].
> Airex PXc®: It is glass fibre reinforced PU foam supplied by Alcan Composites. The
foam exhibits exceptional dimensional stability, chemical and thermal resistance. This
material is available on custom order and hence is very expensive. Also, the minimum
available density of this material is higher than 6 lbs/ft3 density range [45].
> Airex PXw®: It is continuous glass fibre fabric reinforced PU foam supplied by Alcan
composites. The foam is uniquely formulated to exhibit good flexural rigidity on its
own, allowing it to be used with or without face sheets. Airex PXw® also offers
exceptional dimensional stability, chemical and thermal resistance. This material is
available on custom orders and therefore is very expensive. Further, its minimum
available density is larger than 6 lbs/ft3 range [46].
CHAPTER 5. FUSELAGE MATERIAL SELECTION 66
![Page 81: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/81.jpg)
> Foam Filled Honeycomb: Nida Core Corporation and MGI Inc. supply PU foam filled
honeycomb core materials at densities ranging from 5 lbs/ft3 to 20 lbs/ft3. They
combine the properties of honeycomb and foam materials to offer exceptional
compression and shear performance. It is a closed-cell core that can be used with
VARTM and other closed moulding processes. Sealing the honeycomb with foam
imparts increased stand-alone flexural rigidity to this material, which makes it
preferable for mouldless VARTM application. The major drawback is that
manufacturing of PU filled honeycomb is an emerging technology; hence the flexural
rigidity has not yet been quantified. The mechanical properties of 5.56 lbs/ft3 density
material from MGI Inc. were found to be lower than the cross linked PVC cores at the
same density. Also, its machinability, workability and resin absorption characteristics
need to be characterized prior to using them for VARTM applications [47,48].
5.2.4. Core Selection for GeoSurv II Fuselage
The mechanical properties and costs of the structural core materials were considered in order
to select an appropriate core material for mouldless manufacturing. Complete profile of
mechanical properties attributed to various foam and balsa core materials, along with the
supplier details are given in Appendix D. Balsa cores, though exhibiting mechanical properties
superior to foam cores, were not considered in the selection process, as they are too heavy in
their lowest available density. Foam filled honeycombs Airex PXc® and PXw® were also
excluded from the selection process due to limited supply and high costs. This narrowed the list
down to structural foam cores as the primary candidates for the new fuselage.
CHAPTER 5. FUSELAGE MATERIAL SELECTION 67
![Page 82: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/82.jpg)
The structural foams listed in Table 5.2 were assessed for their performance in shear and
compression. This assessment identified PMI (Rohacell), crossed linked PVC (Airex C, Divinycell),
and SAN (Corecell A) as the top three candidates for the new fuselage. Their properties relative
to cost were compared to make the final choice. Since not all of the materials are available at
the same densities, the mechanical properties of the top three foam cores were first
normalized at 4 lbs/ft3 using linear relations. The normalized properties of the top three foam
cores along with their nominal costs are given in Table 5.3. These findings are compared in
Figure 5.4.
Table 5.3: Normalized mechanical properties of the most structural foam cores
Material (4 lbs/ft3)
Rohacell Airex C Corecell A
Compression Strength
(psi)
189 144 56
Percent difference
compared to Corecell A (%)
238 157 0
Shear Strength
(psi)
157 132 89
Percent difference
compared to Corecell A (%)
77 48 0
Cost of 0.5 in thick
material ($/ft2)
11.35 2.50 4.15
Percent difference
compared to Corecell A (%)
173.58 -39.76
0.00
Rohacell (PMI) Airex C (PVC) Corecell (SAN)
a Compression Strength • Shear Strength a Cost
Figure 5.4: Comparison of the selected foam materials at 4 lbs/ft density
CHAPTER 5. FUSELAGE MATERIAL SELECTION 68
![Page 83: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/83.jpg)
The results demonstrate that cross-linked PVC (Airex C) foam cores offer good mechanical
properties at a reasonable cost. Though the latest PMI (Rohacell) foams exhibit 20% better
overall structural performance, their costs are significantly higher than the PVC cores. This
would be a definite concern for the low-cost aspect of this process development work. Hence,
cross-linked PVC foam was chosen as the core material for the new fuselage. Alcan Airex C 70
series structural PVC foam at various density grades, donated by Alcan Composites Inc. was
used throughout this research.
5.3. Selection of Matrix and Reinforcement Materials
The current fuselage (Maley, [3]) was manufactured with SC-780 toughened epoxy matrix
supplied by Applied Poleramic Inc. (API) and AGP-370-5H satin carbon fibre fabric
reinforcement supplied by Hexcel Corporation. The trade studies for the choice of these
materials were carried out as part of the previous mouldless VARTM research and can be found
in [3]. Preliminary Finite Element Analysis (FEA) carried out in [49] showed that most of the
current fuselage structure was overdesigned. In order to reduce the structural weight, Style#
94132-4H satin carbon fibre fabric supplied by BGF industries was selected for the new
fuselage. Additionally, BGF Style# 106-Plain E-glass fabric was chosen as the finishing layer to
provide a smooth surface finish on the outside surfaces of the fuselage. The specifications of
the current and the newly selected fabrics are compared in Table 5.4.
Table 5.4: Carbon fibre fabric specifications
Specification Supplier Fibre Type Tow Weave Style Weight
Hexcel-AGP 370 5H Satin Hexcel Corporation AS4 (Medium Modulus) 6K 5H Satin 11.1 oz/yd2
BGF-94132-4H Satin BGF Industries T300 (Medium Modulus) 4K 4H satin 5.8 oz/yd2
BGF-106-Plain BGF Industries E-glass IK Plain 0.72 oz/yd2
CHAPTER 5. FUSELAGE MATERIAL SELECTION 69
![Page 84: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/84.jpg)
In the new fuselage, PR 2712 infusion epoxy from PTM&W Industries Inc. was substituted for
SC-780 epoxy. Both resin systems are comparable in terms of infusion characteristics and
mechanical properties. The choice was primarily attributed to the lower cost and supply of PR
2712 epoxy from local distributor Composites Canada. The material supplier details for the
matrix and reinforcement materials are given in Appendix A.
5.4. Material Selection for Rigid Inserts
The GeoSurv II fuselage has several structural joints (i.e. bolted pin-joints) through which
discrete loads are introduced into the sandwich structure. Due to the relatively low strength of
the foam core, such locations are susceptible to local failure and must be reinforced with rigid
inserts. The current fuselage uses Delrin® inserts, which were bonded using a structural epoxy
adhesive. Delrin® is machinable and exhibits high specific strength. However, it does not adhere
well to epoxy, often causing failure in the bondline where the insert interfaces with the
fuselage. Hence, alternatives including Fibreglass Reinforced Polymer (FRP), Poly-Ether Ether
Ketone (PEEK™) and glass filled PEEK inserts were considered for the new fuselage. Costs and
mechanical properties of the aforementioned materials are shown in Figure 5.5.
CHAPTER 5. FUSELAGE MATERIAL SELECTION 70
![Page 85: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/85.jpg)
t
500000
450000
400000
350000
300000
250000
200000
150000
100000
50000
0
• Specific Compressive Strength
• Unit Cost
140
120
100
80
60
40
c 3
Fibreglass Reinforced Polymer PEEK 30% GLASS FIUED PEEK DELRIN
Figure 5.5: Comparison of inserts for sandwich assembly
PEEK and Glass Filled PEEK are both considerably lighter than FRP and meet the strength
requirements of the fuselage. However, they are relatively expensive, difficult to machine and
exhibit poor impact characteristics. FRP inserts are relatively inexpensive compared to PEEK and
exhibit good impact properties, specific strength and machinability. Hence, FRP inserts were
selected for the new fuselage. The specifications and supplier information of the inserts are
included in Appendix D.
Note: the unit cost was estimated for 1 in. diameter rods.
CHAPTER 5. FUSELAGE MATERIAL SELECTION 71
![Page 86: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/86.jpg)
CHAPTER 6. FUSELAGE REDESIGN
The redesign objectives and design changes made to the GeoSurv II fuselage are described in
this chapter. An improved GeoSurv II fuselage model is also presented.
6.1. Redesign objectives
The main objective of the fuselage redesign is to improve upon the current design with the
intention of producing a near-net-shape fuselage by mouldless manufacturing. The new
fuselage shall replace the current fuselage in the GeoSurv II prototype, while interfacing with
already manufactured sub-assemblies, such as the wing. The redesign goals and limitations
established based on these intentions are summarized in Table 6.1.
Table 6.1: GeoSurv II fuselage redesign: goals and limitations
Redesign Goals > Reduce weight and improve part quality > Improve the manufacturability of the fuselage
(DFM) > Reduce the steps required to finish the
fuselage: Integrated Manufacturing (IM) > Improve dimensional tolerances (near-net-
shape manufacturing)
Redesign Limitations > Design for mouldless manufacturing: foam
core sandwich construction > The new fuselage shall replace the current
fuselage in the GeoSurv II prototype: No major changes to the Outer Mould Line (OML)
The fuselage redesign work-flow diagram is shown in Figure 6.1. DFM and IM principles were
applied to improve the manufacturability of the current fuselage by mouldless CCBM. The
underlying principle behind this work is continuous product and quality improvement, while
ensuring that the proposed design specifications are in fact achievable with the selected
process. In this work, several DFM and IM principles outlined in [50] were considered, including
minimum part count, ease of fabrication and assembly. The following section presents the
design changes and the improved design concept of the new fuselage.
CHAPTER 6. FUSELAGE REDESIGN 72
![Page 87: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/87.jpg)
Problem Definition Review of the current fuselage
design/manufacturing (Chapter 2)
Selection of New Processing Method and Materials
(Chapters 3-5)
New Fuselage Design (Chapter 6)
Figure 6.1: Fuselage redesign work-flow diagram
6.2. Design Changes to the GeoSurv II Fuselage
The new fuselage walls were designed with edge-stiffened panels in place of the current
'tapered' panels, as shown in Figure 6.2. The required flexural rigidity can be achieved with
thinner edge-stiffened panels than with tapered panels (validated through FEA), resulting in
potential weight reduction and increased internal volume of the fuselage. Additionally, the use
of edge-stiffened panels increases the lengthwise stiffness of the fuselage walls and might
potentially reduce the part distortion experienced during mouldless manufacturing.
CHAPTER 6. FUSELAGE REDESIGN 73
![Page 88: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/88.jpg)
Current Fuselage New Fuselage
Figure 6.2: Current and the new fuselage wall design (units: in.)
In the new design, rigid inserts will be embedded into the foam core prior to fabric layup and
infusion. A bolted sandwich joint that results from this approach is compared to a current
fuselage joint in Figure 6.3. In its current configuration, rigid inserts are secondarily bonded into
the fuselage and washers are sandwiched in-between the nut and bolt. This approach requires
oversized washers to properly transfer out-of-plane loads into the fuselage. The reason is to
minimize stresses within the secondary bond line at the outer surface of the insert. In the new
joint design, the loads are transferred into the fuselage through normal contact of the bolt with
the skin and the insert. Additionally, the bonded area at the skin-insert interface also aids in
distributing the loads into the structure, thereby eliminating the need for oversized washers.
The use of small washers combined with in-situ bonding of skin and inserts is expected to result
in a lighter yet structurally sound bolted joints.
CHAPTER 6. FUSELAGE REDESIGN 74
![Page 89: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/89.jpg)
Bushing [Optional)
Current sandwich bolted New sandwich bolted assembly assembly
Figure 6.3: Current and the new bolted sandwich assembly
The new fuselage also incorporates a landing gear attachment plate to facilitate mounting of
the main landing gear, as shown in Figure 6.4. The new design shifts the attachment point of
the landing gear 5.5 in. aft of the current design. This design change simplifies the complex
swept-back main landing gear design into a straight-beam configuration, as shown in Figure 6.5.
Eliminating the sweep from the landing gear also reduces the moment experienced during one
wheel landing from 12500 in-lbs to 9200 in-lbs. Thus, this design change offers potential weight
savings in the main landing gear strut and the landing gear attachment panel.
CHAPTER 6. FUSELAGE REDESIGN 75
![Page 90: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/90.jpg)
t ^ o )
* \ Plate placement
Landing Gear Attachment Plate
Figure 6.4: New landing gear attachment plate
WL
<—> 6.6 Inches
<-> 3 Inches
Current Landing Gear Configuration New Landing Gear Configuration
Figure 6.5: Current and the new landing gear configurations
Several other design changes were made to improve the manufacturability and minimize the
effort required to prepare the fuselage for assembly. These include: extension of the straight
section of the fuselage wall (Figure 6.6), geometry modification to facilitate mounting of the
shear-pins (Figure 6.7), increased core thickness at the location of the fasteners (Figure 6.8), an
extension on the front bulkhead to substitute for the secondary bonded nosecone bridge
CHAPTER 6. FUSELAGE REDESIGN 76
![Page 91: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/91.jpg)
(Figure 6.9) and reinforcement for mounting of the flight avionics rack (Figure 6.10). In the
process of applying Integrated Manufacturing (IM) principles, the bay separator panel, shown in
Figure 2.5 was not included in the foam components, as it would complicate the vacuum
bagging process. The redesigned fuselage model is shown in Figure 6.11.
m^%^mm
Current Fuselage New Fuselage
Figure 6.6: Fuselage wall straight section extension
Current Design
New Design
Figure 6.7: Design modification at the fairings
CHAPTER 6. FUSELAGE REDESIGN 77
![Page 92: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/92.jpg)
Current Fuselage New Fuselage
Figure 6.8: Increased core thickness at the locations of the fasteners
Current Fuselage New Fuselage
Figure 6.9: Core extension to mount the nosecone
High density (12.5 lbs/ft*) PVC foam reinforcement for
mounting avionics rack
Figure 6.10: Reinforcement for flight avionics rack
CHAPTER 6. FUSELAGE REDESlGf
![Page 93: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/93.jpg)
Current Design New Design
Figure 6.11: Current and the Modified fuselage concept models
CHAPTER 6. FUSELAGE REDESIGN 79
![Page 94: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/94.jpg)
CHAPTER 7. DESIGN OPTIMIZATION
This chapter presents the Finite Element Analysis (FEA) carried out to optimize the new fuselage
and the experimental study conducted to verify selected FEA results.
7.1. GeoSurv II Fuselage FEA
A Finite Element Analysis (FEA) was carried out on the GeoSurv II fuselage, to simulate the
structural performance of the new design under flight manoeuvre and landing loads. The
objective of this FEA was to obtain complete stress-strain profiles at critically loaded areas of
the fuselage, to allow for failure predictions and layup optimization of the new design. The FEA
was carried out using Abaqus 6.8-2, as a linear-elastic stress analysis with several other
simplifying assumptions (discussed in the following sections). The FEA results for selected
loading conditions were verified experimentally using test coupons. The following sections
describe the FEA procedure and results.
7.1.1. FE Model Construction
The parts required to create one half of the fuselage assembly were modelled using the Abaqus
CAE modeller. The parts list includes foam core, skin, wing carry-through spar, bushing, engine
pins, nose and main landing gear pins and wing shear pins. The foam core was modelled as one
3-D deformable solid part and the skin was created from the core, using the "create shell: from
solid" feature available in Abaqus CAE modeller. The carry-through spar was modelled as a
circular shell extrusion and all other components were created as 3-D deformable solids with
appropriate features, as shown in Figure 7.1.
CHAPTER 7. DESIGN OPTIMIZATION 80
![Page 95: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/95.jpg)
The rationale behind the half model was to impose the symmetry condition in the FEA and
hence, reduce the computational time required to run the simulations. To further reduce the
simulation times and expedite the meshing process, several simplifications were made to the
part geometries. The geometric simplifications were as follows:
> Pins were modelled as circular extrusions.
> Fillets and chamfers farther away from the highly loaded areas were excluded from the
part geometry.
> Skins at the top and bottom edges of the fuselage that do not contribute to its load
carrying capabilities were not modelled.
Figure 7.1: Parts modelled for the fuselage FEA
CHAPTER 7. DESIGN OPTIMIZATION 81
![Page 96: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/96.jpg)
7.1.2. Material Properties
The following sections describe the material properties and assumptions employed for various
parts of the fuselage assembly.
Foam Core:
Airex C structural PVC foam at different density grades, supplied by Alcan Composites was used
as the core material for the new fuselage. Previous research in [51] has shown that PVC foams
are transversely isotropic. However, the variation of material properties along different
directions, which is caused by the differences in the cellular development during the foam
expansion process, is rather small and hence ignored for the purpose of this FEA. Thus, the
conditions: Ei=E2=E3 and Gu= G13= G23 were established in the FEA. The values of E and G were
obtained from the material properties published by the manufacturer.
There was no measured or published value of Poisson's ratio (u) for PVC foams. This was
because under uniaxial loading, PVC foams show nearly no deformation along the transverse
axis of the test direction and the respective strains are rather difficult to measure. Additionally,
the PVC foams do not exactly obey the linear-elastic relation between E, G and u. For this
reason, a Poisson's ratio of 0.32 was assumed [52] along with the condition U\i = U13 = 023-
Various density grade PVC foams were modelled with properties as shown in Table 7.1.
Material sections having properties of various density PVC foams were established in the FEA
using the "Engineering Constants" entry, found in the property module of Abaqus CAE.
Subsequently, the parts were assigned appropriate material orientations using local coordinate
systems.
CHAPTER 7. DESIGN OPTIMIZATION 82
![Page 97: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/97.jpg)
The foam core was created as one solid component and was partitioned into different sections.
These partitions were assigned different section properties to represent the parts made with
different density PVC foams (Table 7.1). Thus, a multi-part fuselage foam assembly was created
as single part in the FE model. This section property formulation assumes a perfect bond
between different foam parts, which is a reasonable assumption, given that the adhesive used
to bond the foam parts is stronger than the foam.
Table 7.1: Properties of Airex C PVC foam
Airex C Structural PVC Foam
C70.40
C70. 55
C70. 75
C70.90
C70. 200
Density (lbs/ft3)
2.5
3.7
5.0
6.2
12.5
Ei=E2=E3
(psi) 5947
10000
15080
18850
40600
U l 2 = Ul3 = U23
0.32
0.32
0.32
0.32
0.32
Gl2= Gw= G23 (psi) 1900
3190
4350
5802
10900
Skin:
The new fuselage features carbon-epoxy composite skins with BGF-Style# 94132-4 H satin
carbon fibre fabrics and PTM&W-PR 2712 epoxy. In the FEA, each ply of the woven fabric was
modelled as two unidirectional laminae, with half the total ply thickness. Since no
characterization data were available for these materials at the time of this analysis, the lamina
properties were derived from in-house coupon test data (Table 7.2) available for the current
fuselage materials. The current fuselage skin features Hexcel-AGP-370 5H satin carbon fibre
fabric and API-SC-780 epoxy. Both SC-780 and PR 2712 resin systems have similar mechanical
properties and are compatible with carbon fibre fabrics. The major difference between Style#
CHAPTER 7. DESIGN OPTIMIZATION 83
![Page 98: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/98.jpg)
94132 and AGP 370 fabrics lies in the fibre type and the crimp angle. The AGP 370 fabric
features AS4C type fibres with 6k tow count while the Style# 94132 fabric features T300 fibres
with 3k tows. However, both fabrics are compatible with epoxy resins, which allows for realistic
prediction of lamina properties for the new fabric-resin combination, from the test data
available for the previous reinforcement and matrix materials. The derivation of the new
lamina properties used in the FEA is described in Table 7.3.
Table 7.2: Properties of the current fuselage materials
Material
AGP370 5H satin carbon fibre fabric /SC-780 epoxy
Tensile Strength
(psi)
122400
Tensile Modulus
(psi)
9325900
Strain to Failure
(Tension)
0.013
Shear Strength
(psi)
8600
Shear Modulus
(psi)
102250
Strain to Failure
(In-plane shear)
0.05
CHAPTER 7. DESIGN OPTIMIZATION 84
![Page 99: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/99.jpg)
Table 7.3: Derivation of lamina properties for the new fuselage
Available/Estimated Properties
E 1 = 9325900 psi
E2 = 932590
G12 = 102250
G13 = 10225
G23 = 10225
u = 0.3
Tensile strength = 122400 psi
Strain to failure (Tension) = 0.013
Shear strength = 8600 psi Strain to failure (In-plane shear) = 0.05
Source
In-house coupon testing Hart Smith's 10 % rule [53] In-house coupon testing Hart Smith's 10 % rule [53] Hart Smith's 10 % rule [53] Typical for carbon-epoxy composites In-house coupon testing
In-house coupon testing
In-house coupon testing In-house coupon testing
Knock Down Factor and Rationale
N/A: T 300 and AS4C fibres have similar moduli N/A
N/A: In-plane shear is a matrix dominated property N/A
N/A
N/A
-10% (T300 vs. AS4) +10% (Crimp angle: 3k vs. 6k) -10% (Woven fabric = Unidirectional lamina assumption in the FEA) [54] +10% (T300 vs. AS4) -20% (Woven fabric = Unidirectional lamina assumption in the FEA and additional safety factor of 10%) [54]
15% of the tensile strength; based on in-house test data [3] -20% (Woven fabric = Unidirectional lamina assumption in the FEA and additional safety factor of 10%) [54]
Derived properties
E 1 = 9325900 psi
E2 = 932590 psi
Gi2 = 102250 psi
G13 = 10225 psi
G23 = 10225 psi
u = 0.3
Tensile strength = 101160 psi
Strain to failure (Tension) = 0.011
Shear strength = 15174 psi Strain to failure (In-plane shear) = 0.04
Similar to the foam core, the skin was modelled as a single piece conventional shell in Abaqus
and was partitioned into small sections to facilitate section property assignments and local
mesh refinements. Highly loaded areas of the skin were locally reinforced by assigning different
composite layups.
Carry-through spar, inserts, bushings and pins:
The carry-through spar was assigned conventional shell composite properties (Table 7.2) with
layup [0°/90°]g. The steel pins, Delrin® bushings and fibreglass inserts were modelled as
CHAPTER 7. DESIGN OPTIMIZATION 85
![Page 100: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/100.jpg)
isotropic materials with properties given in Table 7.4. Since the pins and bushings were
employed as connecting elements, to distribute the loads into the structure, an isotropic
property formulation was sufficient for this FEA. The inserts were included in the FE model as
partitioned sections of the foam core, while the pins and bushings were modelled as separate
parts and were assigned their respective material properties. The FEA representation of the
inserts assumes a bondline stronger than the foam, at the insert-foam interface. This
assumption is acceptable, when a structural epoxy adhesive is used for bonding the inserts into
the foam core. Additionally, the inserts are much stronger than the foam and are not expected
to fail before the core or the skin. Hence, to simplify the FEA process, they were also modelled
with isotropic element properties.
Table 7.4: Properties of pins and bushings
Part
Pins Bushings Inserts
Material
Stainless Steel (SS) 316 Delrin®
Fibreglass Rod
Young's Modulus (psi)
27992283 430000 2800000
Poisson's Ratio(u)
0.3 0.3 0.3
7.1.3. Part Meshing Considerations
The foam core, inserts, bushing and pins were meshed with C3D8R elements, which are 8-node
linear bricks with reduced integration and hourglass control. This is a solid, 3-D stress element
in Abaqus that is computationally effective for simplified linear-elastic analysis. The composite
skin around the foam core represents a typical plane-stress scenario, in which the thickness of
the skin is significantly smaller than the other in-plane dimensions. Hence, they were modelled
with S4R elements, which are 4-node, quadrilateral conventional shell elements, featuring
reduced integration, hourglass control and finite membrane strains. It is a common practice to
CHAPTER 7. DESIGN OPTIMIZATION 86
![Page 101: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/101.jpg)
model plane stress states in this fashion due to improved computational efficiency and lack of
through-thickness effects. With the conventional shell formulation, the skin geometry was
modelled as a shell and its thickness was defined by specifying a value and a stacking direction
in the property module of Abaqus. Similar to the skin, the carry-through spar, which features 8
plies of carbon-epoxy composite, was also modelled as a conventional shell using S4R elements.
The fuselage main frame was meshed by creating a series of partitions to sub-divide the
geometry into simple and compatible sections. These partitions (Figure 7.2) facilitate the mesh
propagation in a uniform manner through the entire fuselage geometry. Partitions were
required in the regions where geometric changes or interactions occur. Additionally, the pin
holes were partitioned to be contained in simple square sections, having partitions in a radial
fashion towards the holes. This partitioning strategy, illustrated in Figure 7.3, allows biasing the
mesh seeds and locally refining the mesh near the highly stressed pin-holes. Thus, the areas
away from the critically stressed regions can be meshed in a coarser yet uniform manner to
improve the computational efficiency. All other components of the fuselage assembly were
relatively simple geometries and were meshed by creating partitions, in a similar manner.
CHAPTER 7. DESIGN OPTIMIZATION 87
![Page 102: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/102.jpg)
Figure 7.2: Partitions created on the fuselage skin for meshing
Figure 7.3: Mesh refinement near the pin holes
CHAPTER 7. DESIGN OPTIMIZATION 88
![Page 103: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/103.jpg)
7.1.4. FE Model Assembly and Constraints
The assembly of the fuselage model was created by importing the required part instances into
the Assembly module of Abaqus and applying appropriate position constraints. The final FE
model assembly of the fuselage is shown in Figure 7.4. Following the assembly process, tie
constraints were established at the regions where the pins and the bushings come in contact
with the fuselage structure. In a more realistic attempt to represent bolted pin joints, an analyst
would introduce a contact algorithm in the analysis. This approach is more complex and beyond
the scope of this FEA. In order to keep the tie constraints formulation realistic for bolted joints,
the surface to be tied were carefully chosen based on the loading conditions. Thus, for loading
along a particular direction, as in the case of shear pins, only one half of the pin surface was
tied to the fuselage structure. For all other pins subjected to simultaneous loads in different
directions, the entire pin surface was tied to the fuselage structure, as shown in Figure 7.5.
Figure 7.4: Fuselage FE model assembly
CHAPTER 7. DESIGN OPTIMIZATION 89
![Page 104: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/104.jpg)
Figure 7.5: Tie constraints established between the pins and the fuselage structure
7.1.5. Analysis Steps, Loads and Boundary Conditions
The FEA was implemented in two general static analysis steps: an in-flight step and a landing
step. During the in-flight step, the loads from the wings, engine and payload were applied at
the maximum design ultimate load (DUL) factor of 10.1, which stems from the maximum
positive design limit load (DLL) factor of 6.75 multiplied by the safety factor of 1.5 [55,56].
During the landing step, all in-flight loads were reduced to the maximum landing load factor of
3.5 and the landing loads were introduced. During both analysis steps, the base plate and the
top of surface of the fuselage were fixed and symmetry conditions were applied across the
surface as shown in Figure 7.6. The load cases considered are illustrated in Figure 7.7 to Figure
7.9. The load factors and the loads were obtained from the GeoSurv II design report registry
references [55-57].
CHAPTER 7. DESIGN OPTIMIZATION 90
![Page 105: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/105.jpg)
BC: Fixed- top skin and core
BC: XSYMM: Symmetry about the X= Const, plane
Figure 7.6: Boundary Conditions
Engine Loads:
->Torque at 0.8 RPM 921 in lb/bolt
->Weight 239 lbs 59.75 lbs/bolt
Lift Moment: 8056 in. lbs applied as shear forces at the shear pins
Lift load: 1912 lbs total, applied at the tip of the carry through spar
Figure 7.7: Wing lift and engine loads
CHAPTER 7. DESIGN OPTIMIZATION 91
![Page 106: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/106.jpg)
Main Landing Gear Loads: -^Vertical Landing load: 400 lbs or 100 lbs/in. applied on the pins -^•Side Moment: 9200 in. lbs applied on the area of the washer -^Vertical landing load 600 lbs applied as the pressure load on the area where the landing gear contacts the attachment bracket
Figure 7.8: Main landing gear loads
Air Data boom loads: Weight: 26 lbs Moment: 291 in. lbs
Nose Landing gear loads: Vertical Load: 90 lbs Drag: 54 lbs Side load: 56 lbs
Figure 7.9: Mission avionics and nose landing gear loads
CHAPTER 7. DESIGN OPTIMIZATION 92
![Page 107: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/107.jpg)
7.1.6. Mesh Independence of the Results
A parametric study was performed on the substructure model (Figure 7.10), to determine the
density of the mesh at which the results converge. Throughout the study, the skin and the core
were refined at the same rate, starting from a coarser mesh (1880 total elements), as shown in
Table 7.5. The mesh refinement on the skin is illustrated in Figure 7.11. This study revealed that
the mesh independence of the results with 5% convergence on the Von-Mises stress and 0.2 %
convergence on the displacement occurs at element size of 0.09 in. with the curvature factor of
0.003 in. The corresponding mesh convergence graph is shown in Figure 7.12.
BC: Fixed
Load: Pressure
BC: Symmetry
Figure 7.10: Substructure model used for parametric study
CHAPTER 7. DESIGN OPTIMIZATION 93
![Page 108: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/108.jpg)
Level 1 Level 11
Figure 7.11: Mesh refinement: parametric study level 1 to level 12
Table 7.5: Parametric Study Results
Level
1 2 3 4 5 6 7 8 9 10 11
Element Size/Curvature
control
0.5/0.1 0.25/0.05 0.2/0.01 0.15/0.01
0.125/0.01 0.11/0.01 0.1/0.005 0.1/0.003
0.09/0.003 0.0875/0.001 0.085/0.001
Number of Elements
(n)
1880 5994 13564 23058 36421 52879 72459 85764 111886 146282 154510
Nodal Displacement
(Us)
0.0716217 0.0776539 0.0786073 0.0787259 0.0785697 0.0786482 0.0785664 0.0785424 0.0786624 0.0786709 0.0786694
% Difference
8.08% 1.22% 0.15% 0.20% 0.10% 0.10% 0.03% 0.15% 0.01% 0.00%
Max Von Mises Stress (psi)
8350 9774 7970 7946 11220 9549 10510 10690 10320 10320 10310
% Difference
15.71% 20.33% 0.30% 34.16% 16.09% 9.58% 1.70% 3.52% 0.00% 0.10%
CHAPTER 7. DESIGN OPTIMIZATION 94
![Page 109: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/109.jpg)
11500
11000
10500
=10000
g 8500
8000
7500
7000
+ 5%
20000 40000 60000 80000 100000 120000 140000 160000
Number of Elements (N)
Figure 7.12: Von Mises stress convergence (5%)
Following the convergence study the fuselage was meshed in accordance to the partitioning
and meshing strategies discussed in section 7.1.3. Highly loaded areas were meshed with fine
elements (size 0.09 in. or smaller) and all other regions were meshed in a slightly coarser yet
uniform fashion to improve the computational efficiency. Upon completion, meshes were
verified to correct the elements with poor aspect ratios and internal angles.
7.1.7. FEA Simulations
A series of FEA simulations was executed in order to determine the optimum fuselage layup.
Starting from a base single ply layup, the simulations were iterated with modified layups until
an optimum layup for the fuselage was found. After each simulation, the Von Mises stress, Si2
CHAPTER 7. DESIGN OPTIMIZATION 95
![Page 110: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/110.jpg)
and 13 results were compared against the available material properties, based on which the
layups were modified for the next simulation run.
The final FE model (Figure 7.13) was constructed with 1,213,298 structural elements and
required a minimum of 16 GB computer memory to run the simulation. These simulations were
executed using computer clusters featuring quad-core processors and maximum memory
capacity of 32 GB. At this configuration, each simulation run took approximately 9 hours to
completion.
Figure 7.13: Fuselage FE model
CHAPTER 7. DESIGN OPTIMIZATION 96
![Page 111: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/111.jpg)
7.1.8. FEA Results
The optimized structural foam core components of the new fuselage are shown in Figure 7.14.
Medium to high density PVC foam (5 lbs/ft3 -12.5 lbs/ft3) is used at highly loaded areas such as
the front bulkhead, rear bulkhead, rear walls and the landing gear attachment. The base plate
and the front walls that do not experience high flight manoeuvre or landing loads were
constructed with low density, 3.7 lbs/ft3 PVC foam. The locations of bolted pin-joints were
reinforced with FRP inserts to prevent core crushing under discrete loads.
r Z l 3.7 lbs/ft3
Figure 7.14: Optimized foam core for the new fuselage
CHAPTER 7. DESIGN OPTIMIZATION 97
![Page 112: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/112.jpg)
I I
I
lply[±45°] t= 0.012 in.
2 plies [±45°, 0°/90°] t= 0.024 in.
3 plies [+45°, (0o/90o)2] t= 0.036 in.
4plies[±45°,(0790)2, ±45°,] t= 0.048 in.
8 plies [(±45°)4, (0790°) t= 0.096 in.
Figure 7.15: Optimized skin layup for the new fuselage
Shown in Figure 7.15 is the optimized skin layup for the new fuselage. The locations of pin-
joints and carry-through spar were reinforced with either three or four plies of the carbon fibre
fabric. The landing gear attachment panel was reinforced with eight plies, to allow the structure
to withstand the landing loads. All other regions, those that experience low in-flight and landing
loads were laid-up with one or two plies, depending on the locations and loading conditions.
During the in-flight analysis step, the carry-through spar attachment and the rear bulkhead pin
attachment locations were found to be the highly stressed. Corresponding Von Mises stress and
in-plane shear stress distributions are illustrated in Figure 7.16.
CHAPTER 7. DESIGN OPTIMIZATION 98
![Page 113: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/113.jpg)
The Margin of Safety5 (MS) for each of the Von Mises stress and shear stress conditions was
0.450 and 0.104 respectively. In the landing analysis step, the maximum Von Mises stresses
were experienced at the location where the landing gear attachment bracket interfaces with
the fuselage wall, as shown in Figure 7.17. Its MS was determined to be 0.018, the lowest in the
analysis. The shear stresses experienced by the skin upon landing are shown in Figure 7.18. The
corresponding MS was found to be 0.124. Higher MS values (MS > 0.1) at several locations
suggest that the fuselage layup can be optimized further to reduce weight. However, no more
FEA iterations were performed to bring the MS values down at these locations, in order to leave
provisions for the uncertainties associated with the material property formulations and
modelling assumptions. Full scale destructive testing of the fuselage and further design
optimization were recommended for future research.
In both analysis steps, the stresses in the core were considerably lower than the material
strengths, indicating that core failure is unlikely to occur. This is mainly due to the rigid inserts
distributing the applied discrete loads over a larger surface area within the core. Maximum in-
plane strains on the skin, under in-flight and landing loads were also well below the ultimate
strain of the material (0.011). This showed that the linear-elastic FEA assumption is reasonable.
These results are available in Appendix E.
Margin of Safety (MS) = - 1, is a measure of the structural capacity. MS value of zero means Design Load
that the structure will not take any additional loads before it fails. Design optimization work targets to achieve small positive MS values close to zero. Large positive MS values would mean that the structure is overdesigned and negative MS values would mean that the structure will fail before reaching its maximum design load. (Ref: Burr, A and Cheatham, J: Mechanical Design and Analysis, 2nd edition, section 5.2. Prentice-Hall, 1995)
CHAPTER 7. DESIGN OPTIMIZATION 99
![Page 114: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/114.jpg)
The optimized fuselage structure was estimated to weigh approximately 12.8 lbs. This is 42%
lower compared to its current prototype counterpart, which weighed 22 lbs. The estimates
were based on the surface area and volume of the foam parts used to construct the fuselage
structure. Details on the calculation procedures and the weights estimates of the individual
fuselage panels are also available in Appendix E.
s, Mises SNEG, (fraction = - 1 O). Layer = 1 (Ave;: 7S%) _ - +6.977e+04 B - +3.0OOC+O4
+2.7S0e+04 I Z.SOOc I 04 +2.JS0e+04 + 2 000e+04
I+1.750e+04 +l.sooe+04 + 1.250 e+04 + 1.0OOe+O4 +7.S03e+03 +5.003e+03 +2.503 e+03 +3.615e+00
S, S12 SNEG, (fraction . -1 0), j y e r - 1 (Avg: 75%)
+1.3748+04 +3.0005+0S +2.500S+03 +2.0006+03 +1.5002+03 + i nno*+03 +5.0008+02 +0.0005+00 -5.030e+02 -1.0DOe+03 - l .bjue+iw -2.030e+03 -2.500e+03 -3.030C+03 -1.2S96+04
'• In_Ftight Increment 1: Stap Time => 1.000 Primary var S, Mises
OOft. j Z * . . M M A I W . M . I I I » 1 T". On U 11 U M C O M -
JL i: In Fliqht
Intrenent l:StepTtne = 1X00 Pnmary Var: 3, S12
CMjLeflb Mmi>VEl>*M-i » • » • 6#-I T'.CaZl 11 SI * b><> t*Y«fl'l
Figure 7.16: Von Mises (left) and in-plane shear (right) stresses (psi) in the skin under flight loads
CHAPTER 7. DESIGN OPTIMIZATION 100
![Page 115: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/115.jpg)
3 M^es C1\£G (fraction a (Avg 75%)
+1 C32e+OS • 3 600C+O4
. +2 7S0C+O4 - +2 cQ3e-KS4 • +2 2S9e+04 - *2 C00e+04 -+1 7S0e+O4
•»1 c 0 9 e + 0 4 +J 250e+O4 • l 0306+04 • 7 501B+03
- •SCOle*03 - +2 SOle+03 - +b £33e-01
Figure 7.17: Von Mises stresses (psi) in the skin during the landing step
S, S12 SNEG, (fracaor = (<\vg 75%)
+ 1 350e+D4 +3 O00e+D3 +2 500e+D3 +2 OO0e+D3 +1 SOOfr+33 + 1 O00e+33 + 5 O00e+D2 +O OOOe+DO -5 DOOe-t-02 - I D00e+03 -1 5006+03 -2.D00C+03 -2 500e+03 -3 DOOe-f-03 -1 2S9e-t-04
1 0), La/er = l
Step Lancing Incremant 1 Step Time — 1 0 0 0 Fnman Var S, S12
OMt jZ<ad> A c * q ^ E u ^ 4 4 Vta»o- t 8 -Z T«. M U 13 SJ 54 E i u - D«f I *IT~—-* IMS
Figure 7.18: Shear stresses (psi) in the skin under landing loads
CHAPTER 7. DESIGN OPTIMIZATION 101
![Page 116: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/116.jpg)
7.2. Experimental Verification of the FEA results
An experimental study was carried out to verify the selected FEA results. This was required due
to the simplifying assumptions made in the FEA, as explained in section 7.1. The study aimed to
evaluate two major local loading conditions experienced by the fuselage: in-plane bearing and
out-of-plane bending (Figure 7.19). These conditions were present at the location of shear pin
and on the rear bulkhead. Flat sandwich specimens representing these locations on the
fuselage were loaded until failure and the results were compared with the FEA predictions. The
test matrix and the procedure employed in this study along with the results are discussed in the
following sections.
Rigi
Bolted joint
(a) Bearing load (b) Bending load
Figure 7.19: Loading modes chosen for experiments
CHAPTER 7. DESIGN OPTIMIZATION 102
Sandwich specimen
J insert -mmm-
![Page 117: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/117.jpg)
7.2.1. Test Matrix, Specimen Manufacturing and Test Procedure
The FEA verification test matrix is given in Table 7.6. All specimens were manufactured with
the materials selected for the new fuselage (Chapter 5). The bearing test specimens featured
6.2 lbs/ft3 density PVC foam core having thickness of 0.85 in. and dimensions 6 in. (L) x 6 in.
(W). At the centre of these specimens, the foam core was reinforced with 1 in. diameter FRP
inserts. The specimen thickness was chosen to represent the actual shear pin location in the
new fuselage. Dimensions for the test specimens were chosen based on previous research [58-
60] to avoid edge effects and provide sufficient surface area for clamping of the specimen in the
fixture.
Table 7.6: FEA verification test matrix
Number of specimens
Tested 4 1 3
3 3
Test Type
Bearing Bearing Bearing
Bending Bending
Specimen Dimensions: LxWxT(in.) 6 x 6 x 0.85 6x6x0.85 6 x 6 x 0.85
6x6x0.5 6x6x0.5
Layup
[±45°,0790°, Core]svM [±45°,0790°,0790,Core]SYM [±45°,0790o,0790o,±45°, CoreJsYM
[±45°,0790°, 0790°, Core]SYM [±45°/0790°,0790o
;±45°, Core]SYM
Insert Dimension/Material
l in. diameter GFRP l in. diameter GFRP l in. diameter GFRP
l in. diameter GFRP l in. diameter GFRP
CHAPTER 7. DESIGN OPTIMIZATION 103
![Page 118: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/118.jpg)
y
->X
Z
• > X
Figure 7.20: Geometry of the sandwich specimen
During specimen manufacture, the foam cores were cut into 6 in. x 6 in. pieces. This was
followed by drilling of 1 in. diameter through-hole at the centre of these foam specimens and
bonding the FRP inserts into the holes using structural epoxy adhesive. These foam specimens
were then laid-up with carbon fibre fabric, vacuum bagged and infused with epoxy resin as per
the conventional VARTM method. All specimens were infused under the same vacuum bag
simultaneously, to avoid any inconsistencies in part quality. After curing, the edges of the
specimens were sanded to create smooth, flat edges to facilitate proper installation in the load
frame. A 0.31 in. diameter through-hole was drilled into each specimen to allow insertion of a
tight-fit aluminum bushing. The specimens were then loaded with 0.25 in. diameter steel pins,
CHAPTER 7. DESIGN OPTIMIZATION 104
![Page 119: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/119.jpg)
using a hydraulic Material Testing System (MTS) frame (22 kip capacity, Model 647.10A-01, Serial #:
1305166). The bearing test setup in the load frame is shown in Figure 7.21.
(a) Front view (b) Isometric view
Figure 7.21: Bearing test setup in the load frame
Bending test specimens were manufactured with 0.5 in. thick PVC foam core and loaded at 2.75
in. away from the centerline, using 0.375 in. diameter steel pin to represent the engine pin joint
at the fuselage rear bulkhead. These specimens were prepared for testing by following the
same procedures employed for the bearing specimens. The bending test setup in the load frame
is shown in Figure 7.22. All specimens were loaded at a cross-head displacement rate of 0.19 -^—, mm
while the force and displacement data were collected at 10 Hz.
CHAPTER 7. DESIGN OPTIMIZATION 105
![Page 120: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/120.jpg)
(a) Front view (a) Isometric view
Figure 7.22: Bending test setup in the load frame
7.2.2. FEA Simulations
FE models of the test specimens were created in Abaqus following same procedures employed
in the fuselage FEA. Loads and boundary conditions were applied as shown in Figure 7.23. The
models were meshed uniformly using the optimum element size obtained from the mesh
refinement study. Meshed models are presented in Figure 7.24. First, a range of loads during
which failure might occur, was determined for each layup and specimen type using trial
simulations. This allowed determination of the highest and lowest load to be applied in the
simulations. Then, different FE models were setup for the specimen types discussed in the test
matrix (Table 7.6), each with seven analysis steps. In every analysis step, the applied load was
increased by 50 lbs, starting from the lowest load determined from the trial simulations.
CHAPTER 7. DESIGN OPTIMIZATION 106
![Page 121: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/121.jpg)
Load
(a) Bearing specimen (b) Bending specimen
Figure 7.23: Abaqus model showing the loads and boundary condition of the specimens
(a) Bearing specimen (b) Bendingspecimen
Figure 7.24: Abaqus FE model of the test specimens
CHAPTER 7. DESIGN OPTIMIZATION 107
![Page 122: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/122.jpg)
7.2.3. Results
Force-displacement curves obtained from the bearing test specimens are shown in Figure 7.25.
The curves are shifted along the x-axis for clarity. Similar loading profiles of the specimens
suggested consistent specimen behaviour. The failure load for each specimen tested is given in
Table 7.7. Specimen 7 was discarded as it was an obvious outlier in the group of four-ply
specimens. This was due to misaligned installation of the specimen in the test fixture. This
specimen misalignment caused local crumpling of the specimen edges during the test, leading
to premature failure.
Table 7.7: Bearing test failure loads
Specimen ID
SP#1 SP#2
SP#3
SP#4
SP#5
SP#6 gp ff-j
SP#8
Number of Plies
2 2
2
2
4
4
4
3
Load to Failure (lbs)
983.34
926.65
931.62
971.19
1264.16
1196.14
867.719 1069.49
CHAPTER 7. DESIGN OPTIMIZATION 108
![Page 123: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/123.jpg)
1400
1200
1000
£ 800
o 600
• SP# 1; 2 plies
* SP#5;4plies
• SP#2;2plies
• Sp# 6; 4 plies
SP#3;2plies
^SP#8;3plies
SP#4;2plies
0.05 0.1 0.15 Crosshead Displacement (in)
0.2
Figure 7.25: Bearing test: force-displacement data
Table 7.8 and Figure 7.26 compare average experimental failure loads to those predicted by the
FEA. The experimental failure was defined as the maximum load prior to the abrupt loss of
stiffness in the load-displacement curve. The FEA predicts failure when the in-plane shear stress
on the skin exceeds the material shear strength. The failure modes are compared in Figure 7.27.
For all layups, experimental values were within 10% of the FEA predictions. This correlation
suggests that the assumptions made in the FEA were acceptable for this loading condition.
Table 7.8: Bearing test results: FEA predictions and Experiments
ft of plies
2 3 4
# of specimens tested
4 1 2
Failure load predicted by FEA
(lbs) 983 1154 1202
Average experimental
failure load (lbs) 953 1069 1230
% Difference
3% 8% 2%
CHAPTER 7. DESIGN OPTIMIZATION 109
![Page 124: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/124.jpg)
I f U U -
1350
1300 -
^ 1 2 5 0 " V)
.0 = i 1200 -•o ra O 1150 a> 3 1100 -'ra "" 1050
1000
950
900
•
•
•
\
i i
• Bearing Test FEA
• Experiment
<> I I
i i
2 3 4
Number of Plies
Figure 7.26: Comparison of the failure loads in bearing test
(a) FEA [25x deformation factor] (b) Experiment [Loaded beyondfailure for clarity]
Figure 7.27: Close-up of the bearing failure mode
CHAPTER 7. DESIGN OPTIMIZATION 110
![Page 125: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/125.jpg)
FEA of the bending tests showed failure due to shear stress in the skin, as shown in Figure 7.28.
Predicted failure load of the three-ply and four-ply specimens was 300 lbs and 375 lbs,
respectively. The experiment results were rather interesting. Figure 7.29 shows the force-
displacement curves of the specimens tested. The data obtained for specimen 1 (SP# 1) was
non-linear due to the pin bending and plastically deforming as the applied load was increased.
In second test (SP #2) the pin was reinforced to prevent bending, but the resulting curve was
still non-linear though there was no specimen failure. This time it was due to the fixture
bending and deforming at high loads. Thus, in subsequent tests a reinforced fixture and pin
were used. This test process refinement is evidenced through the increasingly linear slopes SP
#1 to SP# 3. The test setup in SP# 3 was most representative of the actual setup on the rear
bulkhead. The rest of the tests were carried out with this setup (Figure 7.22).
In the bending tests, all specimens carried 600 lbs (1650 in.lbs) without failure, which was well
beyond the loads experienced by the fuselage. This showed that the assumptions made in FEA,
specifically the tie constraints between the pin and the sandwich structure, were too
conservative. Perhaps a better correlation would have been obtained between the FEA
predictions and the experiments if the joint configuration was modelled with proper contact
formulation.
The specimens not failing beyond ultimate loads signifies that the limiting factor in the design
was not the insert or the sandwich structure, but the 3/8" bolt itself. Performing extensive FEA
was not the primary scope of this research. Hence, the specimens were saved for later, more
CHAPTER 7. DESIGN OPTIMIZATION 111
![Page 126: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/126.jpg)
detailed FEA analysis and this research proceeded with manufacturing of the demonstrator
fuselage.
S S I2 SNEG (traction = {AJQ 7 5 M
* 3 0 l o ' O 4 • 1 529*04 *1 260*04 *1 0 l e *04 *7 59o*03
- *5 060*03 *2S3e*03
. *0 0 )*00 - 2 53e>03 - 5 06e*03 -7 59e*03
1019*04 - 1 28a*04 -1 52o*04
• - - 3 20e*O4
ODB Beiw?»ig 4ply oOb Atmqus SlarWatxJ Vwsiot
Slep ln.FlKjW_5 lne*«wn©i« i Step TMPIO - ! 000 Prmnry V I M S S i 2 Defomied Var U Delormabon Seal* Facte*
Figure 7.28: Bending test FEA prediction
• SP# 1; 3 plies • SP#2;3plies - SP#3;3plies
SP#4;4plies < SP#5;4plies *SP#6;4plies
0.2 0.4 0.6 0.8
Crosshead Displacement (in).
1.2
Figure 7.29: Bending test: force-displacement data
CHAPTER 7. DESIGN OPTIMIZATION 112
![Page 127: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/127.jpg)
CHAPTER 8. FUSELAGE MANUFACTURING
This chapter discusses manufacturing of the demonstrator fuselage using mouldless CCBM
method discussed in Chapter 4. The GeoSurv II fuselage, as shown in Figure 6.11, is a large
component with complex geometry, measuring approximately 44 in. x 15 in. x 13 in. Therefore,
any significant error made during manufacturing of the fuselage would result in an expensive
scrap part. Hence, prior to manufacturing of the full scale fuselage, a sample test section was
manufactured to practice and study the mouldless CCBM techniques. This trial component was
sectioned into small pieces and their cross-sections were examined under the microscope to
assess the quality of the CCBM manufacturing. Then, a series of CCBM experiments were
carried out on flat laminates to determine the optimum spacing between the resin distribution
lines in a CIB infusion. Results from these experiments were utilized to develop a conceptual
manufacturing model in Pro/E, followed by the actual fuselage manufacturing. This work is
discussed in the following sections.
8.1. Sample Section Manufacturing
An H-shaped structure representative of the fuselage rear wall assembly was manufactured
using the CCBM method developed for mouldless manufacturing. This was required to assess
the mouldless CCBM and to identify critical areas that may require process refinement. In this
work, the foam parts shown in Figure 8.1 were machined using a 3-axis Computer Numerical
Control (CNC) router. Rabbet features with male-female profiles were integrated into these
parts to ensure precise-fit assembly of the base plate and side walls. The manufactured foam
part (Figure 8.2 (a)) was laid-up with one ply of carbon fibre fabric and a layer of peel ply.
Subsequently, the sandwich layup was wrapped with a layer of static cling PVC (Type I) film
CHAPTER 8. FUSELAGE MANUFACTURING 113
![Page 128: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/128.jpg)
(Translucent, Low Tack, 0.002" Thick, McMaster Carr Inc.). The CCBM bag was manufactured
over this setup, with inlets, outlets and resin distribution lines positioned as shown in Figure 8.2
(b). The PVC film protects the part from silicone contamination during bag manufacturing while
providing a smooth surface finish to the bag. The chemical resistant nature of the PVC film
allowed the bag to be peeled off easily without using any additional release agents. The low
tack adhesive backing of the PVC film eases the wrapping process, while allowing easy removal
from the setup, after bag manufacturing. The CCBM bag was manufactured over this setup,
following the procedure outlined in Appendix B.
The manufactured CCBM bag (Figure 8.2 (c)) featured two resin distribution lines 3 in. apart,
which distributed the resin from the inlet lines located at the top and bottom surfaces of the
base plate into the part. A layer of 1 in. wide Teflease tape was bonded along the perimeter of
the H profile. Once cured, the bag was cut-opened over the taped surface, using an Olfa® utility
knife. The bag was precisely cut, without damaging the fabric reinforcement, in order to retain
the structural integrity of the component. Upon opening and removing the bag, the PVC film
was peeled from the part. This was followed by re-installing the bag over the part and sealing
the split-ends using disposable sealant tape. Then, the inlet lines were clamped-off and the
outlet lines were connected to vacuum ports. Upon vacuum application, it was noticed that the
side walls warped under the applied pressure, due to the unsupported nature of the part
geometry. This problem was solved with the insertion of rigid spacers between the side walls.
Once the setup was complete, resin was introduced into the part via the inlet line. The infusion
of this test component took approximately 6 minutes. Following the infusion, the inlet lines
CHAPTER 8. FUSELAGE MANUFACTURING 114
![Page 129: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/129.jpg)
were attached to vacuum ports to remove any excess resin from the part. The part was cured
under vacuum pressure for 24 hours.
Figure 8.3 shows the cured component before and after removal of the peel ply. There was
some residue of cured resin left in the resin distribution channels and inlet lines, which is
normal in CIB infusion. As the peel ply was removed, the cured resin channels snapped off from
the part, leaving negligible marks on the surface. Significant improvement in the surface quality
was observed around the corner regions, compared to the current fuselage [3]. The part
retained nicely formed corner radii with very few resin starved regions. This improvement in
quality can be attributed to the form-fitted nature of the CCBM bags.
Figure 8.1: Geometry of the test article
In order to assess the quality of the sandwich structure manufactured by the mouldless CCBM
method, the test article was sectioned into small pieces and their cross-sections were examined
under the microscope. In this analysis, images taken from the test specimens were compared
CHAPTER 8. FUSELAGE MANUFACTURING 115
![Page 130: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/130.jpg)
against the images of coupons manufactured using conventional VARTM method. A sample
comparison is shown Figure 8.4. The partially open cells at the skin-core interface were found
to be filled with resin and the trends were comparable between the VARTM and the CCBM
specimens. This is required for good adhesion and proper load transfer between the skin and
the core. More microscopic images of the VARTM and the CCBM specimens are available in
Appendix F.
(a) Sample H-Section (b) CCBM Bag Manufacturing (c) CCBM-CIB Infusion
Figure 8.2: Important features of mouldless CCBM setup
(a) Manufactured test article (b) Manufactured priorto removal of peel ply test article
Figure 8.3: Manufactured component
CHAPTER 8. FUSELAGE MANUFACTURING 116
![Page 131: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/131.jpg)
Figure 8.4: Bondline comparison of VARTM and CCBM manufactured sandwich coupons
8.2.CCBM Experiments and Fuselage Manufacturing Model
In this work, an appropriate spacing between resin distribution lines was determined based on
flat panel experiments. Results from these experiments were combined with the knowledge
acquired from the current prototype manufacturing (Maley, [3]), to develop a conceptual
fuselage manufacturing model. The objective of this work was not shortening the infusion time,
but rather creating a robust and efficient manufacturing setup.
Two flat panel CIB-CCBM experiments were carried out to determine the optimum separation
between the resin distribution lines. Each CCBM bag had a resin inlet, a vacuum outlet and two
resin distribution lines. Important parameters of the test setup are shown in Figure 8.5. Setups
of the two experiments were identical except for the separation between the resin distribution
lines (variable (d) in Figure 8.5 ), which were 6 in. for the first experiment and 10 in. for the
second. The distance between the resin inlet and vacuum outlet was set approximately at 24 in.
This distance was chosen to represent the maximum distance from the centreline of the
fuselage to the centre of the side wall. The resin distribution lines moulded into the bag were
kept 3 in. short of the vacuum outlet line. If this separation was not present, the resin would
CHAPTER 8. FUSELAGE MANUFACTURING 117
![Page 132: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/132.jpg)
travel preferentially from the inlet to outlet rather than infusing the part. The infusions were
carried out on a flat tool surface with four ply carbon fibre fabric preform, at [(±45°)2, (0790°)2]
layup.
The CCBM setup that featured 6 in. separation between the resin distribution lines infused the
part in approximately 13 minutes, while the other setup took 25 minutes for complete infusion.
From manufacturing of the test section, it was found that 3 in. spacing between the resin
distribution lines would require 6 minutes for infusion. From these results, it was decided that
the maximum separation between any two resin distribution lines on the fuselage should be no
more than 8 in. This would result in infusion time of approximately 20 minutes. Though the
infusion time associated with this option was not the shortest, the infusion can be
accomplished with fewer resin distribution lines. This reduces the complexity of the
manufacturing setup and, in turn, makes the additional infusion time tolerable.
Figure 8.5: CCBM experiment setup
CHAPTER 8. FUSELAGE MANUFACTURING 118
![Page 133: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/133.jpg)
Following the experiments, a conceptual CCBM manufacturing model was developed using
Pro/E. The final manufacturing model shown in Figure 8.6 had resin distribution lines at various
spacings along the fuselage walls. Areas of single ply layups had resin distribution lines 8 in.
apart. As the ply count increased towards the rear bulkhead and the front bulkhead of the
fuselage, this spacing was reduced to facilitate simultaneous infusion, as shown in Figure 8.6.
Resin inlet lines were positioned along the centerline of the base plate and over the landing
gear attachment panel, while vacuum outlet lines were placed around the perimeter of the
fuselage. Approximately 2.5 in. distance was kept between the outlet line and the resin
distribution lines.
CHAPTER 8. FUSELAGE MANUFACTURING 119
![Page 134: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/134.jpg)
Isometric View Bottom Section
Figure 8.6: Conceptual CCBM Manufacturing Model
CHAPTER 8. FUSELAGE MANUFACTURING 120
![Page 135: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/135.jpg)
8.3. Fuselage Manufacturing Fuselage manufacturing involved three primary steps: foam core preparation and layup, bag
manufacturing and CCBM infusion. The manufacturing sequence is illustrated in Figure 8.7.
Step 1: Core Preparation/ Layup Step 2: Bag Manufacturing
Manufactured Fuselage Step 3: CCBM Setup/Infusion
Figure 8.7: Mouldless CCBM Process
Manufacturing of the fuselage began with preparation of the foam core. In this process,
required foam parts were machined using a 3 axis-CNC router. The machining process is shown
in Figure 8.8 and the machined parts are shown in Figure 8.9. Then rigid FRP inserts (Figure
8.10), were cut to size from FRP rods, sandblasted and bonded into the foam parts using
PTM&W Inc. ES 6220 epoxy (Pot life 4-6 minutes, cure time 15 minutes). The foam parts were
then bonded together to build the fuselage structure. A custom designed and built wooden
assembly jig, shown in Figure 8.11, was used to hold the foam parts together during the
bonding process. The assembly jig and the alignment features included in the foam parts
(Figure 8.12) served to develop a robust core structure of the fuselage in a cost efficient and
CHAPTER 8. FUSELAGE MANUFACTURING 121
![Page 136: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/136.jpg)
repeatable manner. Following assembly of the foam parts, the fuselage was laid-up with carbon
fibre fabric (BGF Style# 94132), in accordance to the optimized layup determined from the FEA
(Figure 7.15). A layer of thin fibreglass fabric (BGF Style# 106) was laid-up around the outer
surface of the fuselage to provide fine surface finish and thereby minimize the surface
preparation required for painting. The fuselage assembly was wrapped tightly with a layer of
peel ply to create uniform surface finish. During layup, extra care was taken to ensure accurate
fabric orientations and minimum overlaps on the OML. Additionally, it was ensured that all
fabric and peel ply layers were tacked onto the foam core and conformed around the corner
regions. This effort was important to generate good surface finish and minimize resin starved
regions in the fuselage.
Foam Part
Figure 8.8: Machining of the foam parts on the CNC router table
CHAPTER 8. FUSELAGE MANUFACTURING 122
![Page 137: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/137.jpg)
LAvionicsMount Base Plate
<o> ttl
. . • , ! > -
Side Walls * m
Front Walls Front Bulkhead
1 Fairings
:igyr<e 8.9: Feam parts reqyiredl for tfyselag® manyfacSyrifiig
W inserts for ear trough spas' 2*0 g'l V> I
CHAPTER 8. FUSELAGE MANUFACTURING 123
![Page 138: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/138.jpg)
Alignment Fences
Bonding Fixture
Demonstration of the bonding strategy
Figyre 8.11: (Bonding of foam parts in the tenure
v Male-female extrusions
k_ \ ' • L L ^
J. Alignment Shelves
"^'J^| Inserts used as alignment dowels
r^flB Figyre 8.12: Featyres inctoded in tthe foam parts to facilitate precise assembly
CHAPTER 8. FUSELAGE MANUFACTURING
![Page 139: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/139.jpg)
In the next stage of manufacturing, the fuselage core assembly was used as the mould to
manufacture the CCBM bag. The techniques employed in bag fabrication were identical to the
CCBM trial section manufacturing described in section 8.1. The manufactured CCBM bag was
equipped with resin inlets, vacuum outlets and resin distribution channels as illustrated in
Figure 8.6. Upon curing, the bag was cut open along the taped surface and liquid mould release
was sprayed over the inside surface of the bag. This optional step was included in the
manufacturing process to add extra protection to the bag and enhance its self-release
capability.
In the final step, the manufactured CCBM bag was installed back onto the fuselage layup and
sealed with disposable sealant tape. The outlet lines were attached to the vacuum ports. Upon
application of vacuum pressure, the fuselage walls were found to warp inward. The walls were
supported using spacers made from turn buckles with swivel feet (Figure 8.13), to correct this
warpage. Once the setup was complete, a vacuum integrity check was performed, where 1
inHg/min. was considered an acceptable vacuum loss. The CCBM setup had a vacuum loss of
0.3 inHg/min., a significant improvement upon the previous mouldless VARTM setup (Maley,
[3]).
CHAPTER 8. FUSELAGE MANUFACTURING 125
![Page 140: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/140.jpg)
Rigid Spacers
Figure 8.13: Mouldless CCBM setup
After verifying the vacuum integrity of the setup, the fuselage was infused with PT&W 2712
epoxy. Though the infusion time predicted based on the experiments (section 8.2) was 20
minutes, the actual infusion was completed in approximately 45 minutes. This was primarily
due to the resin shortage, which occurred about 10 minutes into the infusion. When this was
noticed, the resin feed lines were clamped off, a new batch of resin was mixed and the infusion
was resumed after 10 minutes. This slowed down the infusion process and resulted in small
resin starved regions in the bottom base plate section, as shown in Figure 8.14. The flow
behaviour during the infusion (Figure 8.7-Step 3) was similar to that observed in the flat panel
CCBM experiments, where the resin travelled preferentially through the distribution lines and
then spread into the fabric.
CHAPTER 8. FUSELAGE MANUFACTURING 126
![Page 141: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/141.jpg)
Figure 8.14: Resin starved regions observed during the infusion
Following the infusion, inlet lines were attached to the vacuum ports to remove any excess
resin from the part. The part was left to cure under vacuum pressure, for 24 hours. The
manufactured fuselage is shown in Figure 8.7.
CHAPTER 8. FUSELAGE MANUFACTURING 127
![Page 142: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/142.jpg)
CHAPTER 9. MANUFACTURING RESULTS
The manufactured fuselage was visually inspected to assess the surface quality. The dimensions
of the new fuselage were measured and compared against the current fuselage manufactured
by mouldless VARTM (Maley, [3]). The weight of the new fuselage was measured and assessed
against the FEA predictions. Finally the PVA was revisited to draw conclusions on the process
viability. This work is described in the following sections.
9.1. Surface Finish, Weight and Tolerances
The manufactured fuselage had a fine peel ply finish with carbon-epoxy skin tightly conformed
around the corners. Some of these surfaces are highlighted in Figure 9.1. With the exception of
the small resin starved regions found in the base plate, the fuselage was fully infused and
showed smooth surface texture. Overall surface finish was better than that observed in
previous manufacturing attempts.
Figure 9.1: Fuselage Manufactured by mouldless CCBM
CHAPTER 9. MANUFACTURING RESULTS 128
![Page 143: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/143.jpg)
There were some issues with the tolerances, although the modified CCBM setup alleviated the
wall distortion observed in the previous mouldless VARTM. The side walls near the top-rear
section of the fuselage were found to be deflected inward, as shown in Figure 9.1. This
deflection was a direct consequence of one turnbuckle spacer falling off, leaving part of the
fuselage walls unsupported during the cure. All other walls remained supported throughout the
entire process. To better characterize the dimensional tolerances, the OML of the
manufactured fuselage was profiled. In this work, the outer dimensions of the fuselage were
measured using rulers, to an accuracy of ±1/16 in. Measurements were taken at a sufficient
number of points to establish an accurate profile of the final shape. Detailed profiling can be
found in Appendix G [61].
Deviations from the target dimensions were below 0.09 in. in most regions, with the exception
of the rear walls on the top half of the 'H' structure and front walls on the bottom half of the 'H'
structure. At these locations the maximum deviations were found to be 0.4 in. and 0.23 in.
respectively. Compared to the previously implemented mouldless VARTM, which resulted in
surface variations of up to 0.25 in. along most fuselage walls, the tolerances achieved with this
manufacturing method better complied with the design targets. This is shown in Table 9.1.
However, there is still a need for improved techniques for controlling part dimensions in
mouldless manufacturing. Better dimension control can perhaps be achieved with the use of a
connected fixture rather than individual spacers (similar to the VARTM setup, [3]). If individual
spacers are used, they need to be embedded into the vacuum bag to ensure more accurate
positioning.
CHAPTER 9. MANUFACTURING RESULTS 129
![Page 144: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/144.jpg)
Table 9.1: Comparison of major dimensions: fuselage design vs. current and new fuselages
Measurement Location
Rear width
Front width
Length (Centerline) -Top
Length (Left) - Top
Length (Right) - Top
Length (Centerline) - Bottom
Length (Left) - Bottom
Length (Right) - Bottom
Length - Fairing to Fairing
Design (in.)
13.34
10.93
43.03
43.08
43.08
44.15
43.02
43.02
15.55
IVIeasured dimensions
(new fuselage) (in.± 1/16 in.)
13.31
10.88
42.97
43.06
43.03
44.19
43.06
43.06
15.47
Deviation in the new fuselage
(in.)
0.03
0.05
0.06
0.02
0.05
-0.04
-0.04 -0.04
0.09
Deviation in the current fuselage
(in.)
-0.06
-0.10
0.20
0.01
0.05
0.31
0.07
0.07
0.05
The manufactured fuselage weighs approximately 14.6 lbs. In order to compensate for the
material to be removed during the installation of the carry-through spar, 0.5 lbs was deducted
from the measured value. Possible weight reduction offered by drilling holes at the locations of
inserts was ignored. Thus the effective weight of the new fuselage was estimated to be 14.1 lbs.
Compared to the current fuselage, the new fuselage offered a total weight saving of 7.9 lbs,
which was approximately 36% of the fuselage weight and 4% of the entire aircraft weight. The
actual weight was close to the theoretical weight estimated using the FEA, which is summarized
in Table 9.2. The Fibre volume fraction (Vf) of the manufactured fuselage was estimated to be
51%. The value of vywas not determined using an ASTM standard, but rather an approximate
value was estimated based on theory, measured weights and assuming zero void content.
Detailed Vf calculations are provided in Appendix H.
CHAPTER 9. MANUFACTURING RESULTS 130
![Page 145: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/145.jpg)
Table 9.2: Comparison of the actual weight and predicted weight of the new fuselage
FUSELAGE WEIGHT SUMMARY Current fuselage weight Predicted weight of the new fuselage based on the FEA Weight of the new fuselage as manufactured
Predicted weight saving
Actual weight saving
22 lbs 12.8 lbs
14.1 lbs 9.2 lbs
(42% of the fuselage, 5% of the aircraft)
7.9 lbs (36% of the fuselage,
4% of the aircraft)
9.2. Process Viability
During manufacturing of the fuselage, it was realized that approximately one week was
required to complete the CCBM bag6. This contradicted the initial PVA estimates, which
accounted only two days for bag manufacturing. Though the CCBM bag could be manufactured
within two days using fast-cure, sprayable CCBM systems, it was in the best interest of this
research to determine whether the CCBM method employed was cost-effective for producing
components in quantities below 10 units. Hence, the PVA was revised to account for a week of
bag manufacturing, and the resulting costs at labour rate of 40$/hr are shown in Figure 9.2.
From these results, it is apparent that the method employed in this research, CCBM III, would
be an attractive choice for manufacturing more than six fuselages. Thus, the process utilized is
viable for mouldless manufacturing of components in low production quantities.
6 Note that in this bag manufacture, some effort was related to the development.
CHAPTER 9. MANUFACTURING RESULTS 131
![Page 146: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/146.jpg)
$5,000
$4,500
$4,000
$3,500
» $3,000 • -to
o $2,500 • -
| $2,000 •
$1,500 -
$1,000 -
$500 • $0
- • - VARTM
-•-CCBM-I
- - C C B M II
- * -CCBM III
4 6
Number of Parts
10
Figure 9.2: Revised PVA cost estimates based on actual labour required for fuselage manufacturing (labour rate $40/hr)
CHAPTER 9. MANUFACTURING RESULTS 132
![Page 147: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/147.jpg)
CHAPTER 10. CONCLUSIONS
The conclusions drawn from this research and recommendations for future work are discussed
in this chapter.
10.1. Conclusions
> Based on an in-depth review of the current LCM processes, CCBM was selected for
mouldless manufacturing of foam-core composite sandwich components. The initial
material and labour cost associated with CCBM is slightly higher than that of
conventional VARTM, but this added cost comes with the benefits of improved process
robustness, repeatability and part quality. A series of flat-panel experiments followed by
a PVA carried out in this research demonstrated that mouldless CCBM with CIB infusion
is a viable option for low volume (<10) production of large, complex components.
> PVA is an effective tool for determining the feasibility of a process. When several
process variants exist, PVA can be used to select the most viable option.
> Effective application of DFM and near-net-shape manufacturing principles improves the
part quality, tolerances and reduces the overall cost of production, in mouldless
manufacturing.
> Simplified FEA techniques are fast and effective in determining the optimum layup of a
composite structure.
> Weight savings of 8 lbs (36% as compared to the current fuselage prototype) were
achieved on the GeoSurv II fuselage using design optimization based on DFM principles
and FEA, supported by coupon tests.
CHAPTER 10. CONCLUSIONS 133
![Page 148: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/148.jpg)
> To facilitate the transfer of discrete loads into a sandwich structure, rigid inserts
embedded into the foam core prior to resin infusion are a robust solution.
Subsequently, holes can be drilled through these local "hard points" and bolted joints
can be created. Such joints are typically lighter and more effective in distributing the
loads into the surrounding structure as compared to those created by bonding rigid
inserts into the manufactured component.
> A new full-scale GeoSurv II fuselage was manufactured to near-net-shape in a single step
infusion using an improved mouldless CCBM method. Although the new fuselage
showed improved surface quality and dimensional tolerances compared to the current
fuselage, some filling and sanding is still required to bring the tolerances closer to the
design specifications. However, achieving near-net-shape tolerances with mouldless
CCBM is not far from reach.
> This research has contributed to several important aspects of low cost composite
structures, including structural design with material selection, structural optimization,
design for manufacture (DFM), process selection, process value analysis (PVA),
manufacturing process planning and development.
10.2. Recommendations for Future Work
The dimensional tolerances of mouldless CCBM manufacturing can be improved by embedding
rigid spacers into the CCBM bag or with the use of accurate fixtures. In the future, more
fuselages will be manufactured using the CCBM bag employed in this research, but with a
modified technique for part shape retention. Results from future manufacturing shall be used
to verify the process repeatability for complex-shape components.
CHAPTER 10. CONCLUSIONS 134
![Page 149: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/149.jpg)
Future manufacturing should also consider using fast-cure, sprayable CCBM systems to improve
the efficiency of bag manufacturing. If not constrained by the design limitations, the foam core
should be designed to withstand the applied vacuum. Thus, the need for additional support
fixtures for part shape retention can be eliminated.
Accurate modelling of the bolted joints using a proper contact algorithm is recommended to
obtain better correlation between FEA results and experiments, particularly for the out-of-
plane bending loads. Additionally, properties of the new matrix and reinforcement materials
should be characterized using coupon testing and the material properties used in the FEA of
this research should be revised to improve the results. Future FEA work should also consider
simulating and optimizing the fuselage under fatigue load and impact loads.
Assumptions used in the FEA were conservative and the MS values were somewhat high at
several locations, which mean that there is still room for optimizing the fuselage layup and
further reducing weight. Future work should consider testing the new fuselage to failure and
optimizing the layup based on the test results and improved FEA methods.
Redesign work in the future should also consider simplifying the geometry to ease the fabric
layup procedures. Additionally, the FRP inserts used in this research have directional properties,
which are undesirable in the presence of multi-axial loads. Thus, future redesign work should
look into isotropic inserts. To further optimize these joints, different insert and joint
configurations should be developed and compared against the current joint design.
CHAPTER 10. CONCLUSIONS 135
![Page 150: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/150.jpg)
Detailed research on flow modelling and developing strategies for optimizing the infusion setup
would further improve the capabilities to manufacture arbitrary (complex) structures using
CCBM.
CHAPTER 10. CONCLUSIONS 136
![Page 151: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/151.jpg)
REFERENCES
1. "Sandwich Panels", Lecture Notes for AERO 4608, Composite Materials, Lecture 14e-Design, Department of Mechanical and Aerospace Engineering, Carleton University, 2008.
2. Diab Inc., Foam Core Materials in the Marine Industry, [Online- Technical Bulletins], 2009, [Cited May 2010] Available: http://www.diaberoup.eom/americas/u literature/u pdf files/u bul pdf/Foam Core Marine TB.pdf.
3. Maley, A.J, "An investigation into low-cost manufacturing of carbon epoxy composites and a novel mouldless technique using the Vacuum Assisted Resin Transfer Moulding (VARTM) method", MASc Thesis, Department of Mechanical and Aerospace Engineering, Carleton University, Ottawa, ON, 2008.
4. Mahendran, M. et al., "Feasibility Study of Closed Cavity Bag Moulding for Novel Mouldless Manufacturing of Carbon-Epoxy Composites", presented at the Canadian Aeronautics and Space Institute AERO '09 Conference- 20th Aerospace Structures and Materials Symposium, Ottawa, ON, 2009.
5. Carleton University Department of Mechanical and Aerospace Engineering and Sander Geophysics Limited. GeoSurv II Unmanned Aerial Vehicle: System Requirements Document (UAV-SRD), Rev. E, 31 March, 2008.
6. Summerscales, J, Searle, T.J, "Low-pressure (vacuum infusion) techniques for moulding large composite structures", Proc. IMechE Vol. 219 Part L: J. Materials: Design and Applications, 07 September 2004.
7. Beckwith, S.W., and Hyland, C.R, "Resin Transfer Moulding: A decade of Technology Advances", SAMPE Journal, Vol. 34, No. 6, November/December, 1998.
8. Rudd, CD. et al., "Liquid Moulding Technologies, Resin Transfer Moulding, Structural Reaction Injection Moulding and Related Processing Techniques", Woodhead Publishing, 1997, pp. 1-35.
9. J.H.A. van der Woude and E.L. Lawton, "Composite Design and Engineering, Fibreglass and Glass Technology", ISBN 978-1-4419-0735-6. Springer Science+Business Media, LLC, 2010, pp. 125-173.
10. Summerscales, J., "Composites Design and Manufacture- Composites Manufacturing Processes" University of Plymouth- Advanced Composites Manufacturing Centre, MATS 324 Lecture Notes, [Online Version Cited May 2010], Available: http://www.tech.plym.ac.uk/sme/MATS324/MATS324C.htm.
11. Potter, K., "Resin Transfer Moulding", Springer-Verlag, 1997.
12. Pant, S. et al., "Characterization of Double-Bagging Effects on 1-D Permeability for Vacuum Assisted Resin Transfer Moulding (VARTM) Process", presented at the Society for the Advancement of Materials and Process Engineering- SAMPE 2010 Conference, Seattle, WA, May 2010.
13. Mahendran, M. et al., "Feasibility Study of Closed Cavity Bag Moulding (CCBM) for Novel Mouldless Manufacturing of Carbon-Epoxy Composites" presented at Canadian Aeronautics and Space Institute AERO'09 Conference, 20th Aerospace Structures and Materials Symposium, Ottawa, ON, 2009.
14. Seemann, WH. "Unitary Vacuum Bag for Forming Fibre Reinforced Composite Articles". US Patent No. 5, 316, 462; May 21,1994.
15. DAI J. et al., "A Comparative Study of Vacuum-Assisted Resin Transfer Molding (VARTM) for Sandwich Panels", Polymer Composites, Vol. 24, No. 6, December 2003.
REFERENCES 137
![Page 152: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/152.jpg)
16 Seemann, WH "Plastic transfer molding techniques for the production of fibre reinforced plastic structures" US Patent No 4,902, 215,1990
17 LI, W etal, "Process and Performance Evaluation of the Vacuum-Assisted Process" Journal of Composite Materials, Vol 38, No 20, 2004
18 Allende, Metal, "Experimental and Numerical Analysis of Flow Behavior in the FASTRAC Liquid Composite Manufacturing Process", Polymer Composites, Vol 25, No 4, August, 2004
19 Niggemann, C et al, "Experimental Investigation of the Controlled Atmospheric Pressure Resin Infusion (CAPRI) Process"', Journal of Composite Materials, 42, SAGE Publications, 2008
20 Takeda, fetal, "Research in the Application of the VaRTM Technique to the Fabrication of Primary Aircraft Composite Structures", Mitsubishi Heavy Industries, Ltd Technical Review Vol 42 No 5, December, 2005
21 Klemeberg, M etal, "Cost Effective CFRP-Fuselage Manufacturing with Liquid Resin Infusion (LRI) -Technologies", Workshop at German Aerospace Centre (DLR) on Final Project of Black Fuselage, Braunschweig, Germany, 2002
22 Louderback et al, "High Performance Infusion System for VARTM Fabrication", U S Patent 6 964 561 B2 November 15, 2005
23 Gibson, R F and Ayormde, E O, "Vibration-Assisted Liquid Composite Moulding", ANTEC Conference Proceedings, Society of Plastics Engineers, Brookfield, CT, vol 2, pp 1544-1547, 2004
24 Arctek Inc, "Closed Cavity Bag Moulding Multi-Port Infusion", CCBM Training Manual, Rev 26 November 1999
25 Fink B K, et al, "Co-Injection Resin Transfer Moulding of Vinyl-Ester and Phenolic Composites", The U S Army Research Laboratory Technical Report, ARL-TR-2150, Accession Number ADA 373528, January 2000
26 Bottler, O et a l , EC-HLM Honeycomb Liquid Moulding, [Online Document] EURO Composites, 2008, [Cited May 2010] Available http //www euro composites com/SiteCollectionDocuments/EC HLM EN pdf
27 Magnum Venus Plastech Ltd , Light RTM (LRTM), [Online], 2007, [Cited September 2010], Available http //www plastech co uk/Mtlrtm html
28 JHM Technologies Inc, RTM Light- The Lower Cost Alternative Takes the Lead in the Closed Moulding Industry, [Online], 2009, [Cited September 2010], Available http //www rtmcomposites com/lrtm html
29 Thagard, J R , et al, "Resin Infusion Between Double Flexible Tooling Evaluation of Process Parameters", Journal of Reinforced Plastics and Composites 23,1767, SAGE Publications, 2004
30 Ruiz, E and Trochu, F, "Flexible Injection A Novel LCM Technology for Low Cost Manufacturing of High Performance Composites Part I Experimental Investigation", 9th International Conference on Flow Processes in Composite Materials (FPCM 9), Montreal, Canada, July 8-10, 2008
31 Oj, B etal, "A Resin Film Infusion Process for Manufacture of Advanced Composite Structures", Composite Structures- 47, pp 471-476,1999
REFERENCES 138
![Page 153: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/153.jpg)
32 Airtech Inc., Vacuum Baggin Techniques, [Online] © 2008, 2009, 2010 - AIRTECH Europe Sari, [Cited September 2010] Available: http //catalogue airtech lu/product php?product id=355
33. Airtech Advanced Materials Group, Airtech Europe S. A., Rubber Silicone Seals, [Online Catalogue], 2008, [Cited 12Jan09], Available: http://cataloeue.airtech.lu/product.php7product id=29&lane=EN.
34. Aaron Miller, CCBM, E-mail communications with Aaron Miller from Composites Canada, Mario Mahendran, Ottawa, 2009.
35. Arctek Inc. CCBM Start-up Kit, [Cited October, 2010], Available: http / / w w w arctekmfusion com/Howtogetstarted htm
36. Crow, K., Value Analysis and Functional Analysis System Technique, DRM Associates, [Online Document], [Cited March 12, 2009], Available: http / / w w w npd-solutions com/va html
37. Lisa Somanchi, et. al., Value Analysis: Overview, The Quality Portal [Online Document], September 09, 2008, [Cited March 12, 2009], Available: http //thequahtyportal com/articles/value htm
38. Diab Inc., Sandwich Concept, Diab Manuals- Diab Sandwich HandbookDiab, [Online Document], [Cited Apr. 2010], Available: h t t p / / w w w diabgroup com/europe/hterature/e pdf files/man pdf/sandwich hb pdf
39. Allen, H.G., "Analysis and Design of Structural Sandwich Panels", Pergamon Press, New York, 1969.
40. Plantema, F. J., "Sandwich construction; the bending and buckling sandwich beams, plates, and shells", Wiley, New York, 1966.
41 . Daniel I.M., et. al., "Major Accomplishments in Composite Materials and Sandwich Structures", An Anthology of ONR Sponsored Research, Springer, New York, 2009.
42. Black S., Getting to the core of Composite Laminates, Composites Technology, Gardner Publications, Inc, [Online Document], 01Oct2003, [Cited 25Jul2008], Available:http / / w w w compositesworld com/articles/getting-to-the-core-of-composite-laminates aspx
43. Nida-Core Corporation, BalsaLite Quality Coated Balsa Core,[Online Document], 2008, [Cited July 2008], Available: http / /www nida-core com/pdfs/pds/nidacore/pds balsalite pdf
44. SP Systems, Core Materials in Polymeric Composites, AZo Journal of Materials Online, AZoM™, [Online Document], 2009, [Cited Aug08], Available: http / / w w w azom com/details asp?ArticlelD=1092
45 Alcan Composites, Foam - AIREX° PXc - Fibre-Reinforced Structural Foam, I.C.A.R.O.H. GmbH/Alcan Composites, [Online Document], 2008, [Cited 27Jul2008], Available: h t t p / / w w w dibond com/alcan/acsites nsf/pages accm3 en/index htmiQpen&p=prod foam pxc&m=4&
type= htm
46. Alcan Composites, Foam - AIREX' PXw - Fibre-Reinforced Structural Foam, I.C.A.R.O.H. GmbH/Alcan Composites, [Online Document], 2008, [Cited 27Jul2008], Available: http / /www dibond com/alcan/acsites nsf/pages accm3 en/index htm |Open&p=prod foam pxw&m=4&
type= htm
47. Nida-Core Corporation, Structural Honeycombs-Foam Filled, [Online Document], 2008, [Cited Aug08], Available: http / / w w w nida-core com/english/mdaprod honey foam htm
139
![Page 154: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/154.jpg)
48. MGI Inc., MiKor Foam Filled Honeycomb, [Online Document], [Cited July 2008], Available: http://www.mgicanada.com/honevcomb.htm.
49. Cochrane D., "Fuselage Finite Element Analysis (FEA)", DR-87-05, Carleton University GeoSurv II Unmanned Aircraft System (UAS) project, 24 March, 2008.
50. Corbett, J.,et. al.. Design for Manufacture: Strategies, Principles and Techniques, Addison-Wesley Publishing Company, Ontario, 1991.
51. Deshpande V.S. and Feleck N.A., "Multi-Axial Yield Behaviour of Polymer Foams", Acta Mater. 49, pp. 1859-1866, 2001.
52. Russell Elkin, PVC Foam Properties, E-mail communications with Russell Elkin, Senior Technical Service Engineer at 3A Composites USA, Mario Mahendran, Ottawa, 2010.
53. Hart-Smith, L. J., 'The ten-percent rule for preliminary sizing of fibrous composite structures", Weight Engineering, vol. 52, no. 2, p. 29-45,1992.
54. Naik R.A., "Failure Analysis of Woven and Braided Fibre Reinforced Composites", NASA Contractor Report 194981, Contract NASI-19708, Notational Aeronautics and Space Administration, Langley Research Centre, Hampton, Virginia, 23681-0001, September 1994.
55. K. Suraweera. V-n Diagram for 150 lb Aircraft and Lift Distribution on a Vertical Tail. Design Report 63-10A. Course AERO 4907. 05 April, 2007.
56. Zakurdaev, A., Update Load Analysis and Methodology, DR 83-02, Carleton University GeoSurv II Unmanned Aircraft System (UAS) project, 31 March, 2008.
57. Buschinelli, M., "Fuselage Loads Summary", DR-117-17, Carleton University GeoSurv II Unmanned Aircraft System (UAS) project, 08 December, 2009.
58. Karakuzu, R. et al., "Failure analysis of woven laminated glass-vinylester composites with pin-loaded hole", Composite Structures, 72; pp. 27-32, 2006.
59. Murat, B. and Sayman, O., Failure analysis of pin-loaded aluminum-glass-epoxy sandwich composite plates, Composites Science and Technology; 63 pp. 727-737, 2003.
60. Baba Okutan, B., Behavior of Pin-loaded Laminated Composites, Experimental Mechanics46: pp. 589-600, 2006.
61. Teutsh, J., "DR137-09: Profiling the Final GeoSurv II Fuselage" Carleton University Aerospace Engineering, Ottawa 26 October 2010.
REFERENCES 140
![Page 155: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/155.jpg)
APPENDICES
Appendix A: Manufacturing Supplies This appendix contains details on the products used in the manufacturing trials and experiments of this research.
Matrix. Reinforcement and Core Materials:
> Infusion Epoxy: PTM&W Inc. PR 2712 [1],[2] > Infusion Epoxy: API SC 780 [3] > Epoxy Adhesive (Fast Cure): PTM&W Inc. ES 6220 [1],[2] > Carbon Fibre Fabric: BGF Style# 94132-3k, 4H satin [1],[4] > Carbon Fibre Fabric: Hexcel AGP 370- 6k, 5H satin [5] > E glass Fabric: BGF Style#106, Ik plain weave [1],[4] > Foam Core: Airex C- structural PVC foam [6] > Inserts: Fibreglass rods [7]
VARTM Supplies:
> Mould Sealant Epoxy - West System Epoxy: Part A105 and Part B 205 or 206 [1],[8] > Paste Mould Release: TR 104 High Temperature [1], [9] > Spray Mould Release: MS-122 AD: PTFE release agent, dry lubricant [10] > Distribution Medium - Resinflow 60 [1],[11] > Vacuum Bag - Strechlon 200, Stretchlon 800 [1], [11] > Sealant Tape - AT-200Y [1], [11] > Breather - Econoweave 44 [1], [11] > Peel Ply - Econoply J [1], [11] > Tubes, Fittings and other consumables- Composites Canada [1], [11]
CCBM Supplies:
> Arctek CCBM system: Progress Plastics and Compounds, Mississauga, Ontario [12] > Teflease tape [1] > 1/8 in. diameter wire wax [13] > Static cling PVC (Type I) film: Translucent, low tack, .002" thick, 12" width, 12'L [7] > Double-sided cloth tape: (Item #. 76125A21- Mc-Master Carr) [7]
Suppliers/Distributors:
[I] Composites Canada, http://www.compositescanada.com/index.php [2] PTM&W Industries Inc., http://www.ptm-w.com/ [3] Applied Poleramic Inc., www.appliedpoleramic.com [4] BGF Industries Inc.. http://www.bgf.com/ [5] Hexcel Corporation, www.Hexcel.com [6] 3a Composites, http://www.corematerials.3acomposites.com/home html?L=l [7] Mc-Master Carr, http //www.mcmaster,com/# [8] West System, www.westsvstem.com [9] TR Industries, http://www.tnndustries.com/ [10] Miller-Stephenson Company Inc, http://www.miller-stephenson.com/ [ I I ] Airtech International, www.airtechonline.com [12] Arctek Inc., http.//www.arctekinfusion.com/What%20is%20CCBM.htm [13] Kindt-Collins Company LLC, http://www.kindt-coilins.com/
Appendix A: Manufacturing Supplies 141
![Page 156: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/156.jpg)
Appendix B: CCBM Bag Manufacturing Procedure
CCBM Manufacturing Trial #1 & 2
1. Prepare the glass tool by cleaning it with Spray-Nine multipurpose cleaner and accurately draw lines to indicate where the seal and resin inlet/outlet lines are to be located ( Figure Bl ) .
Note: The quality of the bag depends on the surface quality of the tool. Fill-in any rough spots or dents on the tool with modelling clay until a smooth surface is achieved.
wn to indicate the t of the seal and resin t lines
Figure B 1: Tool Preparation
2. Apply three coats of TR-104 High Temperature Mould Release Wax over the surface on which the bag is to be fabricated.
Note: Any non-silicone based mould release in liquid or paste form can be used in this step.
3. Slice a 0.5 in. diameter poly tubing in two halves and secure it on the tool using double-sided tape, at the appropriate location of the inlet/outlet lines (Figure B 2).
Flash breaker 1 Tape
Resin inlet/Outlet Lines
Proflex NS Silicone bead
Waxed Glass Tool
Figure B 2: Applying silicone over the prepared tool
Appendix B: CCBM Bag Manufacturing Procedure 142
![Page 157: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/157.jpg)
4. Apply release wax over the poly tubing to ensure easy removal after the bag is cured. 5. Place Airtech Flashbreaker tape upside down along the lines where the bag is to be
sealed, using Airtec 2 fast cure spray adhesive. Note: The use of Airtec 2 spray adhesive to hold the Flashbreaker tape in place contaminated the Flashbreaker tape, causing the sealant tape to permanently adhere to its surface. Hence, in trial # 2, double-sided tape was used to hold the Flashbreaker tape onto the mould.
6. Install 850 g Proflex NS Silicone cartridge in a caulking gun and apply approximately 0.25 in. diameter bead of silicone over the tooling surface, in a pattern as shown in Figure B 2.
7. Using econo-bristle brush or plastic squeegees smear the silicone across the tool along the direction specified in Figure B 2, to create the first layer of silicone.
Note: Care must be taken to brush the silicone over the tool as uniformly as possible. Avoid building up this layer over 0.02 in. thick, as it will take longer to cure.
8. Let the first silicone layer completely cure (approximately 1 hour) in air/moisture and apply another layer of silicone over it. This layer of silicone is to balance out any thickness variations from the first layer and also to build up the thickness of the bag.
9. Once the second layer of silicone is cured, apply a very thin layer of silicone over the bag surface and while the silicone layer is wet tack the Confortex fabric onto the bag as shown in Figure B 3.
Note: The purpose of this layer is to tack the Confortex fabric over the silicone layers. Apply very thin layer of silicone over the entire part and tack the Confortex fabric immediately while the silicone is wet. Overlaps in fabric may be necessary depending on the complexity of the part.
Figure B 3: Laying up the Confortex Reinforcement fabric
10. Once the third layer of silicone (silicone adhesive layer from step 9) is cured, apply a thick layer of silicone over the Confortex fabric and fully impregnate it with silicone. Allow for this layer to cure (approximately 1 hour).
Appendix B: CCBM Bag Manufacturing Procedure 143
![Page 158: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/158.jpg)
11. Apply fourth and fifth layer of silicone over the bag surface, as before, to build up the thickness.
Note: The quality of the bag depends on how uniformly the silicone is brushed each time.
12. Apply another layer of Confortex fabric over the resin inlet and outlet lines and repeat Steps 9 to 11. This step is to locally reinforce the area of inlet and outlet lines to avoid the resin channels collapsing under vacuum pressure.
13. Allow the bag to cure for 24 hours before using it.
Mouldless CCBM Bae Manufacturing: Test Section and Fuselaee
1. Prepare the tool/mould : This process begins with the foam core assembly laid-up with reinforcement.
a. Wrap the entire assembly with a layer of peel ply. b. Wrap the entire assembly with a layer of PVC film (use Teflease tape as necessary
to create a tight wrap around the part). c. Install the resin inlets and outlets onto the setup using double-sided tape. d. Install the resin distribution channels using double-sided tape.
2. Follow the sequence of silicone and Confortex fabric application explained in CCBM bag manufacturing Trial #2.
Note: On large surfaces such as the fuselage walls, apply the silicone over a 2 sq. ft. region at a time. Once the region is fully covered with a layer of silicone, move on to the adjacent region. This is done to apply the silicone while it is wet.
Appendix B: CCBM Bag Manufacturing Procedure 144
![Page 159: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/159.jpg)
Appendix C: Process Value Analysis
Assumptions: * Only the major costs associated with difference in the processing techniques were accounted
in this cost analysis. * Cost of standard laboratory consumables (wax, wipes, paint brushes, squeegees, etc.) were
ignored. * For comparison purposes the labour hours were converted to monetary values at 20$/hr
labour rate. * All prices were in $ CAD and were transferred directly from supplier's or manufacturer's
quotations.
* The prices do not include taxes or shipping/handling costs.
Number of parts Number of Plies
Total Area of Infusion (ft ) Total Resin Weight (lbs)
Seal Perimeter (ft)
Labour Rate ($/hr)
1 2
30 4
10
20
Table C 1: Cost factors for VARTM
Conventional VARTM (Disposable Bagging)
Item
Fabric (including 20% wastage) Resin
Distribution media (Resinflow 60)
Vacuum bag (Stretchlon 200) Peel ply (Econolease) Sealant tape (AT-200Y)
Total Cost Labour hours/part for bagging (hours) Total labour hours for bagging (hours)
Unit cost ($)
21.97/ly 350/ 52 lbs kit
100 ly/$394.30/roll/
100ly:$218.96/ea lOOly: $427.05/roll
4.195$/roll
2.5 hrx3 people
Cost ($)/ unit area (ft2) or length (ft) or weight
(lbs)
$1.83 $6.73
$0.26
$0.15 $0.28 $4.20
Total cost
$131.82 $26.92
$8.58
$12.75 $9.24 $8.39
$197.70
7.50
7.50
Appendix C: Process Value Analysis 145
![Page 160: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/160.jpg)
Table C 2: Cost factors for CCBM I
CCBM (Silicone seal) 1
Item
Fabric (including 20% wastage) Resin
Silicone seal
One time part buildup cost Liquid silicone Distribution media (Resinflow 60) Peel ply (Econolease) Total cost
Labour (hours)
Unit cost ($)
21.97/ ly
350/ 52 lbs kit
9$/ft
33$/cartridge 100 ly/$394.30/roll/ lOOly: $427.05/roll
2 full days
Cost ($)/ unit area (ft2) or length (ft) or weight
(lbs)
$1.83 $6.73
$9.00
$33.25 $0.26 $0.28
16hrs
One time initial cost
Additional cost per part
Total cost
$131.82 $26.92
$90.00
$120.00 $332.50
$8.58 $9.24
$719.06
$542.50
$176.56
Table C 3: Cost factors for CCBM II
CCBM (Sealant tape) II
Item
Fabric (including 20% wastage) Resin
One time part buildup cost
Liquid silicone Distribution media (Resinflow 60) Peel ply (Econolease) Sealant tape Total cost
Labour (hours)
Total labour (hours)
Unit cost ($)
21.97/ ly
350/52 lbs kit
33$/cartridge 100 ly/$394.30/roll/ lOOly: $427.05/roll
4.20$/roll
2- full days
Cost ($)/ unit area (ft2) or length (ft) or weight
(lbs)
$1.83 $6.73
$33.00 $0.26 $0.28 $4.20
15hrs
One time initial cost
Additional cost per part
Total cost
$131.82 $26.92
$100.00
$330.00 $8.58 $9.24 $1.68
$608.24
15.50
$430.00
$178.24
Appendix C: Process Value Analysis 146
![Page 161: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/161.jpg)
Table C 4: Cost factors for CCBM III
CCBM (Integrated with distribution media) III
Item
Fabric (including 20% wastage) Resin
One time part buildup cost
Liquid silicone Sealant tape Total cost Labour (hours) Total labour (hours)
Unit cost ($)
21.97/ ly 350/ 52 lbs kit
33$/cartridge 4.20$ / roll
2- full days
Cost ($)/ unit area (ft2) or length (ft) or weight
(lbs) $1.83 $6.73
$33.00 $4.20
15hrs
One time initial cost
Additional cost per part
Total cost
$131.82 $26.92
$120.00
$330.00 $1.68
$610.42
15.50
$450.00
$160.42
Table C 5: Total cost calculations
Conventional bagging
Cost
$197.70
$395.40
$593.10 $790.80 $988.50
$1,186.20 $1,383.90
$1,581.60
$1,779.30
$1,977.00 $2,174.70 $2,372.40 $2,570.10 $2,767.80 $2,965.50
$3,163.20 $3,360.90 $3,558.60 $3,756.30
$3,954.00
Labour hours
7.50
15.00
22.50 30.00 37.50 45.00 52.50
60.00
67.50
75.00 82.50 90.00 97.50 105.00 112.50
120.00 127.50 135.00 142.50
150.00
CCBMI
Cost
$719.06
$895.62
$1,072.18 $1,248.74 $1,425.30 $1,601.86 $1,778.42
$1,954.98
$2,131.54
$2,308.10 $2,484.66 $2,661.22 $2,837.78 $3,014.34 $3,190.90
$3,367.46 $3,544.02 $3,720.58 $3,897.14
$4,073.70
Labour hours
16.50
17.00
17.50 18.00 18.50 19.00 19.50
20.00
20.50
21.00 21.50 22.00 22.50 23.00 23.50
24.00 24.50 25.00 25.50
26.00
CCBM II
Cost
$608.24
$786.48
$964.72 $1,142.96 $1,321.20 $1,499.44 $1,677.68
$1,855.92
$2,034.16
$2,212.40 $2,390.64
$2,568.88 $2,747.12 $2,925.36 $3,103.60
$3,281.84 $3,460.08 $3,638.32 $3,816.56
$3,994.80
Labour hours
15.50
16.00
16.50 17.00 17.50 18.00 18.50
19.00
19.50
20.00 20.50 21.00 21.50 22.00 22.50
23.00 23.50 24.00 24.50
25.00
CCBM III
Cost
$610.42
$770.84
$931.26 $1,091.68 $1,252.10 $1,412.52 $1,572.94
$1,733.36
$1,893.78
$2,054.20 $2,214.62 $2,375.04 $2,535.46 $2,695.88 $2,856.30
$3,016.72 $3,177.14 $3,337.56 $3,497.98
$3,658.40
Labour hours
16.00
16.40
16.80 17.20 17.60 18.00 18.40
18.80
19.20
19.60 20.00 20.40 20.80 21.20 21.60
22.00 22.40 22.80 23.20
23.60
Part count
1
2
3 4 5 6 7
8
9
10 11 12 13 14 15
16 17 18 19
20
Appendix C: Process Value Analysis 147
![Page 162: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/162.jpg)
Appendix D: Core Materials and Inserts
Figure D 1 to Figure D 6 show the compression, shear and tensile properties of various foam materials utilized in VARTM applications. Their costs are compared in Figure D 7.
600
500
-CELFORT® 300
-KLEGECELLR
-CORECELLT
-AIREX R 63
ROHACELL RIST
AIRCELLT
-DIVINYCELLH
-CORECELLA
-AIREX C 70
AIREXT90
-ROHACELL A
-ELFO
4 5 Density (lbs/ft3)
Figure D1: Compression strengths of foam cores
Appendix D: Core Materials and Inserts 148
![Page 163: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/163.jpg)
35000
30000
«o 25000 D.
I I
-CELFORT® 300
-KLEGECELLR
- CORECELL T
- AIREX R 63
AJRCELLT
-DIVINYCELLH
-CORECELLA
-AIREX C 70
AIREX T 90
-ELFQAM
3 4 5 6 Density (lbs/ft3)
Figure D 2: Compression moduli of foam cores
Figure D 3: Shear strengths of foam cores
Appendix D: Core Materials and Inserts 149
![Page 164: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/164.jpg)
10000
8000
-DIVINYCELLH
-CORECELLA
-AIREX C 70
AIREX T 90
3B
-KLEGECELLR
-CORECELLT
-AIREX R 63
ROHACELL RIST
UT_
3 4 , 5 Density (lbs/ft3)
Figure D 4: Shear moduli of foam cores
800
700
600
i
-DIVINYCELLH
-CORECELLA
-AIREX C 70
ROHACELL RIST
-KLEGECELLR
-CORECELLT
-AIREX R 63
-ROHACELLA
3 4 5 Density (lbs/ft3)
Figure D 5: Tensile strengths of foam cores
Appendix D: Core Materials and Inserts 150
![Page 165: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/165.jpg)
30000
25000
'5 20000 Q.
0)
3 "g 15000
(0
« 10000
5000
3 4 5 6 Density (lbs/ft3)
Figure D 6: Tensile moduli of foam cores
16
14
12 -
10 -
o 'C Q.
6 -
4
2
0 -
1"+! PU
T 1
PET PVC-Cross Linked
SAN PMI
* Prices are for 8 ft x 4 ft, 0.5 in thick sheets, averaged over density range of 3 lbs/ft to 8 lbs/ft * All values are based retail market prices as of Aug08, obtained either from the manufacturers or the
local distributors
Figure D 7: Average prices of various foam cores
Appendix D: Core Materials and Inserts 151
![Page 166: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/166.jpg)
Figure D 8 to Figure D 13 show the properties of balsa cores in comparision to the structural
PMI (Rohacell) foams. The costs of balsa and foam cores are compared in Figure D 14.
4000
3500
s=-3000 a
3)2500 c a> i^
"> 2000 c o » 2 1500 a E O 1000
500
I ^ ^ ROHACELL RIST
^-DIVINYCELLBULSA
-•-BALTEKBULSA
J '/ *
/ /
/ / y
/ *
/
6 8 10 12 Density (lbs/ft3)
14 16
Figure D 8: Compression strengths of balsa cores with reference to PMI foam core
Appendix D: Core Materials and Inserts 152
![Page 167: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/167.jpg)
•a o
in 0) Q. E o o
1000000
900000
800000
700000
600000
500000
400000
300000
200000
100000
0
! I ! -•-CORECELLS
- o - DIVINYCELL BULSA
-•-BALTEKBULSA
•HBB
r?
y / /
/ •
/
6 8 1
Density (lbs/ft3)
0 12 14 16
Figure D 9: Compression moduli of balsa cores with reference to SAN foam core
800
700
600
(0
3 500
O)
c
2 (A w CO
a £
400
300
200
100
-ROHACELL RIST
-DIVINYCELL BULSA
-BALTEKBULSA
6 8 10 12 14 16 Density (lbs/ft3)
Figure D10: Shear strengths of balsa cores with reference to PMI foam core
Appendix D: Core Materials and Inserts 153
![Page 168: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/168.jpg)
<n O.
50000
45000
40000
N35000
'• 30000 w 3 •o 25000 o 2 jj 20000 4) £
w 15000
10000
5000
0
- o - ROHACELL RIST
-^-DIVINYCELL BULSA
-^BALTJEKBULSft
^
O^ ^
0 ^
^
/ / ^
/
6 8 1( Density (lbs/ft3)
12 14 16
Figure D 11: Shear moduli of balsa cores with reference to PMI foam core
4000
3500
3000
0)
S2500
O)
® 2000 CO
« 35 1500 c
1000
500
- ° - ROHACELL RIST
- o - DIVINYCELL BULSA
-•-BALTEKBULSA
~*
/
6 8 10 Density (lbs/ft3)
12 14 16
Figure D12: Tensile strengths of balsa cores with reference to PMI foam core
Appendix D: Core Materials and Inserts 154
![Page 169: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/169.jpg)
900000
800000
700000
•</> 600000 a w • | 500000 •D O S 400000 0)
'55 S 300000
200000
100000
i
-a-ROHACELL RIST
D
^
6 8 10 Density (lbs/ft3)
12 14 16
Figure D 13: Tensile moduli of balsa cores with reference to PMI foam core
O
16
14
12
10
8
6
4
2
0
6
- f c
Foam Cores
d ^
-0F
Balsa Core
Figure D14: Cost of balsa core compared to foam cores
Appendix D: Core Materials and Inserts 155
![Page 170: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/170.jpg)
Material Properties of Core Materials:
1. Divinycell: Diab Corporation http://www.diabgroup.eom/europe/products/e prods l.html
2. Alcan Airex: Alcan Composites http://www.corematerials.3acomposites.com/america.html ?&L=0
3. Nida Core PET, PU, Foam Filled Honey Comb: Nida Core Corporation http://www.nida-core.com/english/index.htm
4. Aircell Polyester Foam Cores-Polyumac Inc. http://www.polyumac.com/Aircellmp.htm 5. Elfoam: Elliott Company- http://www.elliottfoam.com/tech.html 6. Corecell: SP Systems- http://www.marineware.com/ccp 2.asp 7. Rohacell: Evonik Industries /Degussa
http://www.rohacell.com/en/performanceplastics8344.html 8. Last-A-Foam: General Plastics Inc.
http://www.generalplastics.com/products/product detail.php?pid=19
Material Properties of Inserts:
1. FRP inserts, PEEK, Glass Filled Peek: Mc-Master Carr, http://www.mcmaster.eom/#
Appendix D: Core Materials and Inserts 156
![Page 171: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/171.jpg)
Appendix E: FEA Results and Weight Estimates
FEA Results:
Table E 1: FEA Results: In-Flight Analysis Step
Analysis Step
In-Flight
In-Flight
In-Flight
In-Flight
Location
Carry-through Spar
Shear Pin
Rear Bulkhead
Front Bulkhead
Max. Von Mises Stress
(psi) 60,000
40, 000
69,000
20,000
Max. In-Plane Shear Stress,
S12 (psi)
13,000
6,500
9,500
4,000
Max In-Plane Strain. (IE)
0.010
0.004
0.008
0.003
Table E 2: FEA Results: Landing analysis step
Analysis Step
Landing
Landing
Landing
Landing
Landing
Location
Carry-through Spar
Shear Pin
Rear Bulkhead
Front Bulkhead
Landing Gear Attachment
Max. Von Mises Stress
(psi)
34, 000
10, 000
30, 000
28, 000
52,000
Max. In-Plane Shear Stress,
S« (psi)
3,500
2,500
4,000
7,000
13,500
Max In-Plane Strain. (13)
0.007
0.002
0.003
0.005
0.011
Weight Estimates:
Table E 3: Weights of fuselage panels predicted by the FEA
*(lbs)*
Skin Core
Base plate
0.79 0.16
Side walls
1.54 0.67
Fairings
0.29 0.73
Front bulkhead
0.29 0.17
Rear bulkhead
0.44 0.31
Bushings
1.71 N/A
Landing gear
bracket 1.99 0.42
Nose cone
bridge 0.15 0.07
Rear side walls 0.92 0.46
Bay separator
panel 0.17 0.05
Appendix E: FEA Results and Weight Estimates 157
![Page 172: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/172.jpg)
Appendix F: Microscopic Image Analysis
VARTM Specimens:
Good = Partially open cells at the skin-core interface are filled with resin.
Bad = Voids at the skin core interface.
Figure F1: Sample cross-sections of VARTM specimens
Appendix F: Microscopic Image Analysis 158
![Page 173: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/173.jpg)
CCBM Specimens:
Figure F 2: Images of the CCBM specimens at random locations
Appendix F: Microscopic Image Analysis 159
![Page 174: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/174.jpg)
Appendix G: Fuselage Profiling
The outer mould line (OML) (Figure G 1) of the new fuselage was measured around all of the
edges with particular attention to the regions that interface with the access panels. The entire
process and results are discussed in the following few sections.
\- 20 81
Figure G 1: Overall dimensions of the fuselage assembly
Appendix G: Fuselage Profiling 160
![Page 175: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/175.jpg)
Procedure:
The profiling of the fuselage was carried out in a process involving several steps. These steps
are outlined below.
1. The Pro/E model of the assembled fuselage (without the carbon fibre skin) was examined and all of the points at which measurements would be taken were determined.
2. The dimensions of the Pro/E model at each of these points were recorded. Experimental skin thickness was added to the Pro/E dimensions to obtain the ideal dimensions listed in Table G 1 through Table G 6.
3. The fuselage was marked with masking tape at each of the points where measurements would be taken, as can be seen in Figure G 2 and Figure G 3.
4. All of the measurements of the fuselage were taken with a shop ruler to a precision of at least ±1/16".
5. All of the data was inserted into a MS Excel™ template to obtain the deviations found in Table G 1 through Table G 6 and to determine the tolerances for each set of results.
6. The results were used to sketch the actual profiles of the fuselage on top of the ideal profiles as shown in Figure G 4 and Figure G 5.
Appendix G: Fuselage Profiling 161
![Page 176: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/176.jpg)
> T3 T3 <T> Q_ x '
c (/> ro_
era
3 era
NOTES ' Measurements A through AP are taken normal to the planes they are shown on
' Points A and X are 1 5" frorr. the rear of the rear buikheac • Points A-D and W-T are measured in 5" increments from the rear oT the rear bulkhead
' Points G-J and O-R ere measured in 5" increments from the front bulkhead along ther respective edges
' Points AE and AK align with point I * Measurements AS and AT are lire put equvalenls tg they are shown on measurements AW and AX
* _ / W_7 v / U__/ T_j !_§ R_/ Q / CD M
»
![Page 177: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/177.jpg)
Figure G 3: Marking the fuselage with masking tape to identify the points shown in Figure G 2
Results:
Table G 1 through Table G 6 show the measured values at the locations illustrated in Figure G 2. The results are split into separate tables according to which part of the fuselage they pertain to. The deviation columns are colour coded by the magnitude of the deviation with the darker red showing the greatest deviation and the brighter green being showing the smallest.
Note that the fuselage widths along the top and the bottom surfaces have the greatest deviations. This is of most interest for the Structures Fuselage Team, since these are the areas that pertain to the attachment of the access panels. Figure G 4 and Figure G 5 show precisely where these measurements were taken and also the show spline curves (in red) connecting where the measurements were taken, in order to give the complete shape profile of the actual fuselage. Note that in creating these spline curves it was assumed that the deviation is symmetric about the centreline of the fuselage. This assumption is not perfect, but is adequate for the purpose of demonstrating the deviation.
Appendix G: Fuselage Profiling 163
![Page 178: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/178.jpg)
Table G 1: Width results across the top of the fuselage
Measurement ID
A B C D E F G H 1 J K
Measurement ID
0 p Q R S T U V
w X
Measurement [in.]
(±1/16)
13 8/32 13 0/32 12 30/32 13 0/32 13 4/32 13 12/32 13 2/32 12 20/32 12 5/32 11 21/32 117/32
Table G 2: Width results
Measurement [in.]
(+1/16) 11 8/32 11 20/32 12 8/32 12 30/32 13 14/32 13 10/32 13 14/32 13 10/32 13 14/32 13 13/32
Ideal [in.] A-F (±0.010) G-K (±0.006)
13.34 13.34 13.34 13.34 13.34 13.34 13.22 12.75 12.28 11.81 11.34
Deviation [in.] A-F (±0.073) G-K (±0.069)
-0.09 -0.34 -0.40 -0.34 -0.21 0.04 -0.16 -0.12 -0.12 -0.16 -0.12
across the bottom of the fuselage
Ideal [in.] (±0.006)
11.33 11.88 12.36 12.83
13.34+0.010 13.34±0.010 13.38±0.018 13.38±0.018 13.48±0.034 13.48±0.034
Deviation [in.] (±0.069)
-0.08 -0.26 -0.11 0.11
0.10±0.073 -0.02+0.073 0.05+0.081 -0.07+0.081 -0.04+0.097 -0.07+0.097
Appendix G: Fuselage Profiling 164
![Page 179: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/179.jpg)
22.19
1.5
K 11-7/32 J
11-21/32
I 12-5/32
H 12-5/8
G 13-1/16
F 13-3/8
E 13-1/8
D 43-
C 12-15/16
B 13
A 13-1/4
5.0
- M 0 . 0
15.0
20.0
5
/
20.0
15.0
T 5.0
J_
10.0
Figure G 4: Outline of top of fuselage showing the actual (in red) and the ideal profiles and the locations of the measurement points shown in Table G 1. The dimensions shown are the measured dimensions. The positions of the measurements are
shown relative to the fuselage Outer Mould Line (OML).
Appendix G: Fuselage Profiling 165
![Page 180: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/180.jpg)
22.19
Figure G 5: Outline of bottom of fuselage showing the actual (in red) and the ideal profiles and the locations of the measurement points shown in Table G 2. The dimensions shown are the measured dimensions. The positions of the
measurements are shown relative to the fuselage Outer Mould Line (OML).
Appendix G: Fuselage Profiling 166
![Page 181: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/181.jpg)
Table G 3: Front bulkhead height and width dimensions
Measurement ID Measurement [in.]
(+1/16)
Ideal [in.] (±0.01)
10.93 10.93 10.93 14.22 15.10 14.22
Deviation [in.] (±0.073)
-0.05 -0.05 -0.05 -0.06 -0.03 -0.06
L M N
AG AH Al
10 28/32 10 28/32 10 28/32 14 5/32 15 2/32 14 5/32
Table G 4: Rear bulkhead height and width dimensions
Measurement ID Measurement [in.]
(±1/16)
Ideal [in.] Deviation [in.]
Y Z
AA AO AP
13 12/32 13 10/32 13 10/32 15 16/32 15 16/32
13.50+0.034 13.34+0.014 13.34+0.014 15.52+0.022 15.52+0.022
-0.13+0.097 -0.05+0.077 -0.05+0.077 -0.02+0.085 -0.02+0.085
Table G 5: Front bulkhead tab dimensions
Measurement ID Measurement [in.] (±0.001)
Ideal [in.] (±0.01)
Deviation [in.] (±0.011)
AY
AZ BA BB BC
4.012 0.907
0.386 0.374 0.512
4.035
0.895 0.355
0.355 0.500+0.00
-0.044 -0.009
0.010 -0.002 0.012
Table G 6: Fuselage length results
Measurement ID Measurement [in.]
(± 1/16)
Ideal [in.] Deviation [in.]
AQ AR AS AT AU AV AW AX
22 5/16 20 13/16 22 5/16
20 13/16 22 4/16
20 13/16 22 5/16
20 13/16
22.23+0.007 20.97+0.005 22.29+0.017 20.82+0.005 22.24+0.007
20.97+0.005 22.29+0.017 20.82+0.005
0.07+0.070 -0.16+0.068 0.02+0.080 -0.01+0.068 0.01+0.070 -0.16+0.068 0.02+0.080 0.07+0.068
Appendix G: Fuselage Profiling 167
![Page 182: AN IMPROVED MOULDLESS MANUFACTURING METHOD FOR …€¦ · Value Analysis (PVA). Subsequently, the fuselage was redesigned, for improved manufacturability. The new design was optimized](https://reader034.vdocuments.net/reader034/viewer/2022051607/6038361cfc9cc82b4d5aa727/html5/thumbnails/182.jpg)
Appendix H: Fibre Volume Fraction Calculation
Weight of resin infused (lbs)
Likewise,
Weight of fabric on the fuselage (lbs)=
Pi 300 fibres
V. skin
2.916 lbs_
i3
0.0425 ft3
68.64 —j, where pn ft
M.
K,
4.884
110.6 Ibs^
ft' 0.0442 ft3
Using, Vresm+Vfabnc = Vskin, (zero void content is assumed, and weave effects of the fabric is
ignored)
Vf ~ 51%.
Appendix H: Fibre Volume Fraction Calculation 168